Module Awso_rekognition_asyncSource

Provides the S3 bucket name and object name. The region for the S3 bucket containing the S3 object must match the region you use for Amazon Rekognition operations. For Amazon Rekognition to process an S3 object, the user must have permission to access the S3 object. For more information, see How Amazon Rekognition works with IAM in the Amazon Rekognition Developer Guide.

A description of the dominant colors in an image.

Identifies the bounding box around the label, face, text, object of interest, or personal protective equipment. The left (x-coordinate) and top (y-coordinate) are coordinates representing the top and left sides of the bounding box. Note that the upper-left corner of the image is the origin (0,0). The top and left values returned are ratios of the overall image size. For example, if the input image is 700x200 pixels, and the top-left coordinate of the bounding box is 350x50 pixels, the API returns a left value of 0.5 (350/700) and a top value of 0.25 (50/200). The width and height values represent the dimensions of the bounding box as a ratio of the overall image dimension. For example, if the input image is 700x200 pixels, and the bounding box width is 70 pixels, the width returned is 0.1. The bounding box coordinates can have negative values. For example, if Amazon Rekognition is able to detect a face that is at the image edge and is only partially visible, the service can return coordinates that are outside the image bounds and, depending on the image edge, you might get negative values or values greater than 1 for the left or top values.

Information about an item of Personal Protective Equipment covering a corresponding body part. For more information, see DetectProtectiveEquipment.

The S3 bucket that contains an Amazon Sagemaker Ground Truth format manifest file.

The X and Y coordinates of a point on an image or video frame. The X and Y values are ratios of the overall image size or video resolution. For example, if an input image is 700x200 and the values are X=0.5 and Y=0.25, then the point is at the (350,50) pixel coordinate on the image. An array of Point objects makes up a Polygon. A Polygon is returned by DetectText and by DetectCustomLabels Polygon represents a fine-grained polygon around a detected item. For more information, see Geometry in the Amazon Rekognition Developer Guide.

The API returns a prediction of an emotion based on a person's facial expressions, along with the confidence level for the predicted emotion. It is not a determination of the person’s internal emotional state and should not be used in such a way. For example, a person pretending to have a sad face might not be sad emotionally. The API is not intended to be used, and you may not use it, in a manner that violates the EU Artificial Intelligence Act or any other applicable law.

Indicates the location of the landmark on the face.

Information about an item of Personal Protective Equipment (PPE) detected by DetectProtectiveEquipment. For more information, see DetectProtectiveEquipment.

Assets are the images that you use to train and evaluate a model version. Assets can also contain validation information that you use to debug a failed model training.

Structure containing the estimated age range, in years, for a face. Amazon Rekognition estimates an age range for faces detected in the input image. Estimated age ranges can overlap. A face of a 5-year-old might have an estimated range of 4-6, while the face of a 6-year-old might have an estimated range of 4-8.

Indicates whether or not the face has a beard, and the confidence level in the determination.

Indicates the direction the eyes are gazing in (independent of the head pose) as determined by its pitch and yaw.

Indicates whether or not the eyes on the face are open, and the confidence level in the determination.

Indicates whether or not the face is wearing eye glasses, and the confidence level in the determination.

FaceOccluded should return "true" with a high confidence score if a detected face’s eyes, nose, and mouth are partially captured or if they are covered by masks, dark sunglasses, cell phones, hands, or other objects. FaceOccluded should return "false" with a high confidence score if common occurrences that do not impact face verification are detected, such as eye glasses, lightly tinted sunglasses, strands of hair, and others. You can use FaceOccluded to determine if an obstruction on a face negatively impacts using the image for face matching.

The predicted gender of a detected face. Amazon Rekognition makes gender binary (male/female) predictions based on the physical appearance of a face in a particular image. This kind of prediction is not designed to categorize a person’s gender identity, and you shouldn't use Amazon Rekognition to make such a determination. For example, a male actor wearing a long-haired wig and earrings for a role might be predicted as female. Using Amazon Rekognition to make gender binary predictions is best suited for use cases where aggregate gender distribution statistics need to be analyzed without identifying specific users. For example, the percentage of female users compared to male users on a social media platform. We don't recommend using gender binary predictions to make decisions that impact an individual's rights, privacy, or access to services.

Identifies face image brightness and sharpness.

Indicates whether or not the mouth on the face is open, and the confidence level in the determination.

Indicates whether or not the face has a mustache, and the confidence level in the determination.

Indicates the pose of the face as determined by its pitch, roll, and yaw.

Indicates whether or not the face is smiling, and the confidence level in the determination.

Indicates whether or not the face is wearing sunglasses, and the confidence level in the determination.

An instance of a label returned by Amazon Rekognition Image (DetectLabels) or by Amazon Rekognition Video (GetLabelDetection).

A potential alias of for a given label.

The category that applies to a given label.

A parent label for a label. A label can have 0, 1, or more parents.

Describes the face properties such as the bounding box, face ID, image ID of the input image, and external image ID that you assigned.

Configuration for Moderation Labels Detection.

Object containing information about the model versions of selected features in a given job.

Information about where an object (DetectCustomLabels) or text (DetectText) is located on an image.

Structure containing attributes of the face that the algorithm detected. A FaceDetail object contains either the default facial attributes or all facial attributes. The default attributes are BoundingBox, Confidence, Landmarks, Pose, and Quality. GetFaceDetection is the only Amazon Rekognition Video stored video operation that can return a FaceDetail object with all attributes. To specify which attributes to return, use the FaceAttributes input parameter for StartFaceDetection. The following Amazon Rekognition Video operations return only the default attributes. The corresponding Start operations don't have a FaceAttributes input parameter: GetCelebrityRecognition GetPersonTracking GetFaceSearch The Amazon Rekognition Image DetectFaces and IndexFaces operations can return all facial attributes. To specify which attributes to return, use the Attributes input parameter for DetectFaces. For IndexFaces, use the DetectAttributes input parameter.

Provides face metadata. In addition, it also provides the confidence in the match of this face with the input face.

Contains information regarding the confidence and name of a detected content type.

The known gender identity for the celebrity that matches the provided ID. The known gender identity can be Male, Female, Nonbinary, or Unlisted.

Information about a body part detected by DetectProtectiveEquipment that contains PPE. An array of ProtectiveEquipmentBodyPart objects is returned for each person detected by DetectProtectiveEquipment.

Summary information for an Amazon Rekognition Custom Labels dataset. For more information, see ProjectDescription.

Configuration options for Content Moderation training.

The S3 bucket that contains the training summary. The training summary includes aggregated evaluation metrics for the entire testing dataset and metrics for each individual label. You get the training summary S3 bucket location by calling DescribeProjectVersions.

The dataset used for testing. Optionally, if AutoCreate is set, Amazon Rekognition uses the training dataset to create a test dataset with a temporary split of the training dataset.

Contains the Amazon S3 bucket location of the validation data for a model training job. The validation data includes error information for individual JSON Lines in the dataset. For more information, see Debugging a Failed Model Training in the Amazon Rekognition Custom Labels Developer Guide. You get the ValidationData object for the training dataset (TrainingDataResult) and the test dataset (TestingDataResult) by calling DescribeProjectVersions. The assets array contains a single Asset object. The GroundTruthManifest field of the Asset object contains the S3 bucket location of the validation data.

The dataset used for training.

Object specifying the acceptable range of challenge versions.

Specifies a location within the frame that Rekognition checks for objects of interest such as text, labels, or faces. It uses a BoundingBox or Polygon to set a region of the screen. A word, face, or label is included in the region if it is more than half in that region. If there is more than one region, the word, face, or label is compared with all regions of the screen. Any object of interest that is more than half in a region is kept in the results.

A filter that allows you to control the black frame detection by specifying the black levels and pixel coverage of black pixels in a frame. As videos can come from multiple sources, formats, and time periods, they may contain different standards and varying noise levels for black frames that need to be accounted for. For more information, see StartSegmentDetection.

Contains metadata for a UserID matched with a given face.

Provides face metadata for target image faces that are analyzed by CompareFaces and RecognizeCelebrities.

Contains input information for a media analysis job.

Details about the error that resulted in failure of the job.

Summary that provides statistics on input manifest and errors identified in the input manifest.

Configuration options for a media analysis job. Configuration is operation-specific.

Output configuration provided in the job creation request.

Contains the results for a media analysis job created with StartMediaAnalysisJob.

Statistics about a label used in a dataset. For more information, see DatasetLabelDescription.

Information about a word or line of text detected by DetectText. The DetectedText field contains the text that Amazon Rekognition detected in the image. Every word and line has an identifier (Id). Each word belongs to a line and has a parent identifier (ParentId) that identifies the line of text in which the word appears. The word Id is also an index for the word within a line of words. For more information, see Detecting text in the Amazon Rekognition Developer Guide.

Information about a shot detection segment detected in a video. For more information, see SegmentDetection.

Information about a technical cue segment. For more information, see SegmentDetection.

Details about a person detected in a video analysis request.

Structure containing details about the detected label, including the name, detected instances, parent labels, and level of confidence.

Provides information about a single type of inappropriate, unwanted, or offensive content found in an image or video. Each type of moderated content has a label within a hierarchical taxonomy. For more information, see Content moderation in the Amazon Rekognition Developer Guide.

Information about a recognized celebrity.

The quality of an image provided for label detection, with regard to brightness, sharpness, and contrast.

Feature specific configuration for the training job. Configuration provided for the job must match the feature type parameter associated with project. If configuration and feature type do not match an InvalidParameterException is returned.

The evaluation results for the training of a model.

The S3 bucket and folder location where training output is placed.

Sagemaker Groundtruth format manifest files for the input, output and validation datasets that are used and created during testing.

The data validation manifest created for the training dataset during model training.

An ordered list of preferred challenge type and versions.

The label detection settings you want to use in your stream processor. This includes the labels you want the stream processor to detect and the minimum confidence level allowed to label objects.

A set of parameters that allow you to filter out certain results from your returned results.

Specifies the starting point in a Kinesis stream to start processing. You can use the producer timestamp or the fragment number. One of either producer timestamp or fragment number is required. If you use the producer timestamp, you must put the time in milliseconds. For more information about fragment numbers, see Fragment.

Filters for the shot detection segments returned by GetSegmentDetection. For more information, see StartSegmentDetectionFilters.

Filters for the technical segments returned by GetSegmentDetection. For more information, see StartSegmentDetectionFilters.

Contains filters for the object labels returned by DetectLabels. Filters can be inclusive, exclusive, or a combination of both and can be applied to individual labels or entire label categories. To see a list of label categories, see Detecting Labels.

Provides UserID metadata along with the confidence in the match of this UserID with the input face.

Face details inferred from the image but not used for search. The response attribute contains reasons for why a face wasn't used for Search.

Provides information about a celebrity recognized by the RecognizeCelebrities operation.

Metadata of the user stored in a collection.

An object that recognizes faces or labels in a streaming video. An Amazon Rekognition stream processor is created by a call to CreateStreamProcessor. The request parameters for CreateStreamProcessor describe the Kinesis video stream source for the streaming video, face recognition parameters, and where to stream the analysis resullts.

Describes a project policy in the response from ListProjectPolicies.

Description for a media analysis job.

Describes a dataset label. For more information, see ListDatasetLabels.

Object containing both the face metadata (stored in the backend database), and facial attributes that are detected but aren't stored in the database.

A face that IndexFaces detected, but didn't index. Use the Reasons response attribute to determine why a face wasn't indexed.

Information about text detected in a video. Incudes the detected text, the time in milliseconds from the start of the video that the text was detected, and where it was detected on the screen.

Metadata information about an audio stream. An array of AudioMetadata objects for the audio streams found in a stored video is returned by GetSegmentDetection.

A technical cue or shot detection segment detected in a video. An array of SegmentDetection objects containing all segments detected in a stored video is returned by GetSegmentDetection.

Information about the type of a segment requested in a call to StartSegmentDetection. An array of SegmentTypeInfo objects is returned by the response from GetSegmentDetection.

Information about a video that Amazon Rekognition analyzed. Videometadata is returned in every page of paginated responses from a Amazon Rekognition video operation.

Details and path tracking information for a single time a person's path is tracked in a video. Amazon Rekognition operations that track people's paths return an array of PersonDetection objects with elements for each time a person's path is tracked in a video. For more information, see GetPersonTracking in the Amazon Rekognition Developer Guide.

Information about a label detected in a video analysis request and the time the label was detected in the video.

Information about a person whose face matches a face(s) in an Amazon Rekognition collection. Includes information about the faces in the Amazon Rekognition collection (FaceMatch), information about the person (PersonDetail), and the time stamp for when the person was detected in a video. An array of PersonMatch objects is returned by GetFaceSearch.

An image that is picked from the Face Liveness video and returned for audit trail purposes, returned as Base64-encoded bytes.

Information about a face detected in a video analysis request and the time the face was detected in the video.

Information about an inappropriate, unwanted, or offensive content label detection in a stored video.

Information about a detected celebrity and the time the celebrity was detected in a stored video. For more information, see GetCelebrityRecognition in the Amazon Rekognition Developer Guide.

A training dataset or a test dataset used in a dataset distribution operation. For more information, see DistributeDatasetEntries.

Provides face metadata for the faces that are disassociated from a specific UserID.

Contains metadata like FaceId, UserID, and Reasons, for a face that was unsuccessfully disassociated.

A person detected by a call to DetectProtectiveEquipment. The API returns all persons detected in the input image in an array of ProtectiveEquipmentPerson objects.

Allows you to set attributes of the image. Currently, you can declare an image as free of personally identifiable information.

The background of the image with regard to image quality and dominant colors.

The foreground of the image with regard to image quality and dominant colors.

Settings for the IMAGE_PROPERTIES feature type.

A custom label detected in an image by a call to DetectCustomLabels.

Kinesis video stream stream that provides the source streaming video for a Amazon Rekognition Video stream processor. For more information, see CreateStreamProcessor in the Amazon Rekognition Developer Guide.

The Kinesis data stream Amazon Rekognition to which the analysis results of a Amazon Rekognition stream processor are streamed. For more information, see CreateStreamProcessor in the Amazon Rekognition Developer Guide.

The Amazon S3 bucket location to which Amazon Rekognition publishes the detailed inference results of a video analysis operation. These results include the name of the stream processor resource, the session ID of the stream processing session, and labeled timestamps and bounding boxes for detected labels.

Label detection settings to use on a streaming video. Defining the settings is required in the request parameter for CreateStreamProcessor. Including this setting in the CreateStreamProcessor request enables you to use the stream processor for label detection. You can then select what you want the stream processor to detect, such as people or pets. When the stream processor has started, one notification is sent for each object class specified. For example, if packages and pets are selected, one SNS notification is published the first time a package is detected and one SNS notification is published the first time a pet is detected, as well as an end-of-session summary.

Input face recognition parameters for an Amazon Rekognition stream processor. Includes the collection to use for face recognition and the face attributes to detect. Defining the settings is required in the request parameter for CreateStreamProcessor.

A description of an Amazon Rekognition Custom Labels project. For more information, see DescribeProjects.

A description of a version of a Amazon Rekognition project version.

Provides statistics about a dataset. For more information, see DescribeDataset.

Contains metadata like FaceId, UserID, and Reasons, for a face that was unsuccessfully deleted.

Contains settings that specify the location of an Amazon S3 bucket used to store the output of a Face Liveness session. Note that the S3 bucket must be located in the caller's AWS account and in the same region as the Face Liveness end-point. Additionally, the Amazon S3 object keys are auto-generated by the Face Liveness system.

Provides information about a face in a target image that matches the source image face analyzed by CompareFaces. The Face property contains the bounding box of the face in the target image. The Similarity property is the confidence that the source image face matches the face in the bounding box.

Provides face metadata for the faces that are associated to a specific UserID.

Contains metadata like FaceId, UserID, and Reasons, for a face that was unsuccessfully associated.

You are not authorized to perform the action.

Amazon Rekognition experienced a service issue. Try your call again.

Input parameter violated a constraint. Validate your parameter before calling the API operation again.

The number of requests exceeded your throughput limit. If you want to increase this limit, contact Amazon Rekognition.

The specified resource is already being used.

The resource specified in the request cannot be found.

Amazon Rekognition is temporarily unable to process the request. Try your call again.

Allows you to opt in or opt out to share data with Rekognition to improve model performance. You can choose this option at the account level or on a per-stream basis. Note that if you opt out at the account level this setting is ignored on individual streams.

The stream processor settings that you want to update. ConnectedHome settings can be updated to detect different labels with a different minimum confidence.

An Amazon Rekognition service limit was exceeded. For example, if you start too many jobs concurrently, subsequent calls to start operations (ex: StartLabelDetection) will raise a LimitExceededException exception (HTTP status code: 400) until the number of concurrently running jobs is below the Amazon Rekognition service limit.

Describes updates or additions to a dataset. A Single update or addition is an entry (JSON Line) that provides information about a single image. To update an existing entry, you match the source-ref field of the update entry with the source-ref filed of the entry that you want to update. If the source-ref field doesn't match an existing entry, the entry is added to dataset as a new entry.

The size of the collection exceeds the allowed limit. For more information, see Guidelines and quotas in Amazon Rekognition in the Amazon Rekognition Developer Guide.

A ClientRequestToken input parameter was reused with an operation, but at least one of the other input parameters is different from the previous call to the operation.

Amazon Rekognition is unable to access the S3 object specified in the request.

The file size or duration of the supplied media is too large. The maximum file size is 10GB. The maximum duration is 6 hours.

The Amazon Simple Notification Service topic to which Amazon Rekognition publishes the completion status of a video analysis operation. For more information, see Calling Amazon Rekognition Video operations. Note that the Amazon SNS topic must have a topic name that begins with AmazonRekognition if you are using the AmazonRekognitionServiceRole permissions policy to access the topic. For more information, see Giving access to multiple Amazon SNS topics.

Set of optional parameters that let you set the criteria text must meet to be included in your response. WordFilter looks at a word's height, width and minimum confidence. RegionOfInterest lets you set a specific region of the screen to look for text in.

Video file stored in an Amazon S3 bucket. Amazon Rekognition video start operations such as StartLabelDetection use Video to specify a video for analysis. The supported file formats are .mp4, .mov and .avi.

This is a required parameter for label detection stream processors and should not be used to start a face search stream processor.

Specifies when to stop processing the stream. You can specify a maximum amount of time to process the video.

Filters applied to the technical cue or shot detection segments. For more information, see StartSegmentDetection.

Indicates that a provided manifest file is empty or larger than the allowed limit.

The requested resource isn't ready. For example, this exception occurs when you call DetectCustomLabels with a model version that isn't deployed.

Contains the specified filters that should be applied to a list of returned GENERAL_LABELS.

Provides face metadata such as FaceId, BoundingBox, Confidence of the input face used for search.

Contains metadata about a User searched for within a collection.

The input image size exceeds the allowed limit. If you are calling DetectProtectiveEquipment, the image size or resolution exceeds the allowed limit. For more information, see Guidelines and quotas in Amazon Rekognition in the Amazon Rekognition Developer Guide.

The provided image format is not supported.

Contains data regarding the input face used for a search.

Provides the input image either as bytes or an S3 object. You pass image bytes to an Amazon Rekognition API operation by using the Bytes property. For example, you would use the Bytes property to pass an image loaded from a local file system. Image bytes passed by using the Bytes property must be base64-encoded. Your code may not need to encode image bytes if you are using an AWS SDK to call Amazon Rekognition API operations. For more information, see Analyzing an Image Loaded from a Local File System in the Amazon Rekognition Developer Guide. You pass images stored in an S3 bucket to an Amazon Rekognition API operation by using the S3Object property. Images stored in an S3 bucket do not need to be base64-encoded. The region for the S3 bucket containing the S3 object must match the region you use for Amazon Rekognition operations. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes using the Bytes property is not supported. You must first upload the image to an Amazon S3 bucket and then call the operation using the S3Object property. For Amazon Rekognition to process an S3 object, the user must have permission to access the S3 object. For more information, see How Amazon Rekognition works with IAM in the Amazon Rekognition Developer Guide.

The supplied revision id for the project policy is invalid.

The format of the project policy document that you supplied to PutProjectPolicy is incorrect.

A resource with the specified ID already exists.

Pagination token in the request is not valid.

Contains metadata about a label detection request, including the SortBy and AggregateBy options.

Describes the type and version of the challenge being used for the Face Liveness session.

Occurs when a given sessionId is not found.

Contains metadata about a content moderation request, including the SortBy and AggregateBy options.

A User with the same Id already exists within the collection, or the update or deletion of the User caused an inconsistent state. **

A set of optional parameters that you can use to set the criteria that the text must meet to be included in your response. WordFilter looks at a word’s height, width, and minimum confidence. RegionOfInterest lets you set a specific region of the image to look for text in.

Summary information for required items of personal protective equipment (PPE) detected on persons by a call to DetectProtectiveEquipment. You specify the required type of PPE in the SummarizationAttributes (ProtectiveEquipmentSummarizationAttributes) input parameter. The summary includes which persons were detected wearing the required personal protective equipment (PersonsWithRequiredEquipment), which persons were detected as not wearing the required PPE (PersonsWithoutRequiredEquipment), and the persons in which a determination could not be made (PersonsIndeterminate). To get a total for each category, use the size of the field array. For example, to find out how many people were detected as wearing the specified PPE, use the size of the PersonsWithRequiredEquipment array. If you want to find out more about a person, such as the location (BoundingBox) of the person on the image, use the person ID in each array element. Each person ID matches the ID field of a ProtectiveEquipmentPerson object returned in the Persons array by DetectProtectiveEquipment.

Specifies summary attributes to return from a call to DetectProtectiveEquipment. You can specify which types of PPE to summarize. You can also specify a minimum confidence value for detections. Summary information is returned in the Summary (ProtectiveEquipmentSummary) field of the response from DetectProtectiveEquipment. The summary includes which persons in an image were detected wearing the requested types of person protective equipment (PPE), which persons were detected as not wearing PPE, and the persons in which a determination could not be made. For more information, see ProtectiveEquipmentSummary.

Shows the results of the human in the loop evaluation. If there is no HumanLoopArn, the input did not trigger human review.

The number of in-progress human reviews you have has exceeded the number allowed.

Sets up the flow definition the image will be sent to if one of the conditions is met. You can also set certain attributes of the image before review.

Information about the quality and dominant colors of an input image. Quality and color information is returned for the entire image, foreground, and background.

Settings for the DetectLabels request. Settings can include filters for both GENERAL_LABELS and IMAGE_PROPERTIES. GENERAL_LABELS filters can be inclusive or exclusive and applied to individual labels or label categories. IMAGE_PROPERTIES filters allow specification of a maximum number of dominant colors.

Information about the source streaming video.

The Amazon Simple Notification Service topic to which Amazon Rekognition publishes the object detection results and completion status of a video analysis operation. Amazon Rekognition publishes a notification the first time an object of interest or a person is detected in the video stream. For example, if Amazon Rekognition detects a person at second 2, a pet at second 4, and a person again at second 5, Amazon Rekognition sends 2 object class detected notifications, one for a person at second 2 and one for a pet at second 4. Amazon Rekognition also publishes an an end-of-session notification with a summary when the stream processing session is complete.

Information about the Amazon Kinesis Data Streams stream to which a Amazon Rekognition Video stream processor streams the results of a video analysis. For more information, see CreateStreamProcessor in the Amazon Rekognition Developer Guide.

Input parameters used in a streaming video analyzed by a Amazon Rekognition stream processor. You can use FaceSearch to recognize faces in a streaming video, or you can use ConnectedHome to detect labels.

A description for a dataset. For more information, see DescribeDataset. The status fields Status, StatusMessage, and StatusMessageCode reflect the last operation on the dataset.

A session settings object. It contains settings for the operation to be performed. It accepts arguments for OutputConfig and AuditImagesLimit.

The source that Amazon Rekognition Custom Labels uses to create a dataset. To use an Amazon Sagemaker format manifest file, specify the S3 bucket location in the GroundTruthManifest field. The S3 bucket must be in your AWS account. To create a copy of an existing dataset, specify the Amazon Resource Name (ARN) of an existing dataset in DatasetArn. You need to specify a value for DatasetArn or GroundTruthManifest, but not both. if you supply both values, or if you don't specify any values, an InvalidParameterException exception occurs. For more information, see CreateDataset.

Type that describes the face Amazon Rekognition chose to compare with the faces in the target. This contains a bounding box for the selected face and confidence level that the bounding box contains a face. Note that Amazon Rekognition selects the largest face in the source image for this comparison.

Allows you to update a stream processor. You can change some settings and regions of interest and delete certain parameters.

Allows you to update a stream processor. You can change some settings and regions of interest and delete certain parameters.

This operation applies only to Amazon Rekognition Custom Labels. Adds or updates one or more entries (images) in a dataset. An entry is a JSON Line which contains the information for a single image, including the image location, assigned labels, and object location bounding boxes. For more information, see Image-Level labels in manifest files and Object localization in manifest files in the Amazon Rekognition Custom Labels Developer Guide. If the source-ref field in the JSON line references an existing image, the existing image in the dataset is updated. If source-ref field doesn't reference an existing image, the image is added as a new image to the dataset. You specify the changes that you want to make in the Changes input parameter. There isn't a limit to the number JSON Lines that you can change, but the size of Changes must be less than 5MB. UpdateDatasetEntries returns immediatly, but the dataset update might take a while to complete. Use DescribeDataset to check the current status. The dataset updated successfully if the value of Status is UPDATE_COMPLETE. To check if any non-terminal errors occured, call ListDatasetEntries and check for the presence of errors lists in the JSON Lines. Dataset update fails if a terminal error occurs (Status = UPDATE_FAILED). Currently, you can't access the terminal error information from the Amazon Rekognition Custom Labels SDK. This operation requires permissions to perform the rekognition:UpdateDatasetEntries action.

This operation applies only to Amazon Rekognition Custom Labels. Adds or updates one or more entries (images) in a dataset. An entry is a JSON Line which contains the information for a single image, including the image location, assigned labels, and object location bounding boxes. For more information, see Image-Level labels in manifest files and Object localization in manifest files in the Amazon Rekognition Custom Labels Developer Guide. If the source-ref field in the JSON line references an existing image, the existing image in the dataset is updated. If source-ref field doesn't reference an existing image, the image is added as a new image to the dataset. You specify the changes that you want to make in the Changes input parameter. There isn't a limit to the number JSON Lines that you can change, but the size of Changes must be less than 5MB. UpdateDatasetEntries returns immediatly, but the dataset update might take a while to complete. Use DescribeDataset to check the current status. The dataset updated successfully if the value of Status is UPDATE_COMPLETE. To check if any non-terminal errors occured, call ListDatasetEntries and check for the presence of errors lists in the JSON Lines. Dataset update fails if a terminal error occurs (Status = UPDATE_FAILED). Currently, you can't access the terminal error information from the Amazon Rekognition Custom Labels SDK. This operation requires permissions to perform the rekognition:UpdateDatasetEntries action.

Removes one or more tags from an Amazon Rekognition collection, stream processor, or Custom Labels model. This operation requires permissions to perform the rekognition:UntagResource action.

Removes one or more tags from an Amazon Rekognition collection, stream processor, or Custom Labels model. This operation requires permissions to perform the rekognition:UntagResource action.

Adds one or more key-value tags to an Amazon Rekognition collection, stream processor, or Custom Labels model. For more information, see Tagging AWS Resources. This operation requires permissions to perform the rekognition:TagResource action.

Adds one or more key-value tags to an Amazon Rekognition collection, stream processor, or Custom Labels model. For more information, see Tagging AWS Resources. This operation requires permissions to perform the rekognition:TagResource action.

Stops a running stream processor that was created by CreateStreamProcessor.

Stops a running stream processor that was created by CreateStreamProcessor.

This operation applies only to Amazon Rekognition Custom Labels. Stops a running model. The operation might take a while to complete. To check the current status, call DescribeProjectVersions. Only applies to Custom Labels projects. This operation requires permissions to perform the rekognition:StopProjectVersion action.

This operation applies only to Amazon Rekognition Custom Labels. Stops a running model. The operation might take a while to complete. To check the current status, call DescribeProjectVersions. Only applies to Custom Labels projects. This operation requires permissions to perform the rekognition:StopProjectVersion action.

Starts asynchronous detection of text in a stored video. Amazon Rekognition Video can detect text in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartTextDetection returns a job identifier (JobId) which you use to get the results of the operation. When text detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the text detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetTextDetection and pass the job identifier (JobId) from the initial call to StartTextDetection.

Starts asynchronous detection of text in a stored video. Amazon Rekognition Video can detect text in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartTextDetection returns a job identifier (JobId) which you use to get the results of the operation. When text detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the text detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetTextDetection and pass the job identifier (JobId) from the initial call to StartTextDetection.

Starts processing a stream processor. You create a stream processor by calling CreateStreamProcessor. To tell StartStreamProcessor which stream processor to start, use the value of the Name field specified in the call to CreateStreamProcessor. If you are using a label detection stream processor to detect labels, you need to provide a Start selector and a Stop selector to determine the length of the stream processing time.

Starts processing a stream processor. You create a stream processor by calling CreateStreamProcessor. To tell StartStreamProcessor which stream processor to start, use the value of the Name field specified in the call to CreateStreamProcessor. If you are using a label detection stream processor to detect labels, you need to provide a Start selector and a Stop selector to determine the length of the stream processing time.

Starts asynchronous detection of segment detection in a stored video. Amazon Rekognition Video can detect segments in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartSegmentDetection returns a job identifier (JobId) which you use to get the results of the operation. When segment detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. You can use the Filters (StartSegmentDetectionFilters) input parameter to specify the minimum detection confidence returned in the response. Within Filters, use ShotFilter (StartShotDetectionFilter) to filter detected shots. Use TechnicalCueFilter (StartTechnicalCueDetectionFilter) to filter technical cues. To get the results of the segment detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetSegmentDetection and pass the job identifier (JobId) from the initial call to StartSegmentDetection. For more information, see Detecting video segments in stored video in the Amazon Rekognition Developer Guide.

Starts asynchronous detection of segment detection in a stored video. Amazon Rekognition Video can detect segments in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartSegmentDetection returns a job identifier (JobId) which you use to get the results of the operation. When segment detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. You can use the Filters (StartSegmentDetectionFilters) input parameter to specify the minimum detection confidence returned in the response. Within Filters, use ShotFilter (StartShotDetectionFilter) to filter detected shots. Use TechnicalCueFilter (StartTechnicalCueDetectionFilter) to filter technical cues. To get the results of the segment detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetSegmentDetection and pass the job identifier (JobId) from the initial call to StartSegmentDetection. For more information, see Detecting video segments in stored video in the Amazon Rekognition Developer Guide.

This operation applies only to Amazon Rekognition Custom Labels. Starts the running of the version of a model. Starting a model takes a while to complete. To check the current state of the model, use DescribeProjectVersions. Once the model is running, you can detect custom labels in new images by calling DetectCustomLabels. You are charged for the amount of time that the model is running. To stop a running model, call StopProjectVersion. This operation requires permissions to perform the rekognition:StartProjectVersion action.

This operation applies only to Amazon Rekognition Custom Labels. Starts the running of the version of a model. Starting a model takes a while to complete. To check the current state of the model, use DescribeProjectVersions. Once the model is running, you can detect custom labels in new images by calling DetectCustomLabels. You are charged for the amount of time that the model is running. To stop a running model, call StopProjectVersion. This operation requires permissions to perform the rekognition:StartProjectVersion action.

End of support notice: On October 31, 2025, AWS will discontinue support for Amazon Rekognition People Pathing. After October 31, 2025, you will no longer be able to use the Rekognition People Pathing capability. For more information, visit this blog post. Starts the asynchronous tracking of a person's path in a stored video. Amazon Rekognition Video can track the path of people in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartPersonTracking returns a job identifier (JobId) which you use to get the results of the operation. When label detection is finished, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the person detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetPersonTracking and pass the job identifier (JobId) from the initial call to StartPersonTracking.

End of support notice: On October 31, 2025, AWS will discontinue support for Amazon Rekognition People Pathing. After October 31, 2025, you will no longer be able to use the Rekognition People Pathing capability. For more information, visit this blog post. Starts the asynchronous tracking of a person's path in a stored video. Amazon Rekognition Video can track the path of people in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartPersonTracking returns a job identifier (JobId) which you use to get the results of the operation. When label detection is finished, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the person detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetPersonTracking and pass the job identifier (JobId) from the initial call to StartPersonTracking.

Initiates a new media analysis job. Accepts a manifest file in an Amazon S3 bucket. The output is a manifest file and a summary of the manifest stored in the Amazon S3 bucket.

Initiates a new media analysis job. Accepts a manifest file in an Amazon S3 bucket. The output is a manifest file and a summary of the manifest stored in the Amazon S3 bucket.

Starts asynchronous detection of labels in a stored video. Amazon Rekognition Video can detect labels in a video. Labels are instances of real-world entities. This includes objects like flower, tree, and table; events like wedding, graduation, and birthday party; concepts like landscape, evening, and nature; and activities like a person getting out of a car or a person skiing. The video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartLabelDetection returns a job identifier (JobId) which you use to get the results of the operation. When label detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the label detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetLabelDetection and pass the job identifier (JobId) from the initial call to StartLabelDetection. Optional Parameters StartLabelDetection has the GENERAL_LABELS Feature applied by default. This feature allows you to provide filtering criteria to the Settings parameter. You can filter with sets of individual labels or with label categories. You can specify inclusive filters, exclusive filters, or a combination of inclusive and exclusive filters. For more information on filtering, see Detecting labels in a video. You can specify MinConfidence to control the confidence threshold for the labels returned. The default is 50.

Starts asynchronous detection of labels in a stored video. Amazon Rekognition Video can detect labels in a video. Labels are instances of real-world entities. This includes objects like flower, tree, and table; events like wedding, graduation, and birthday party; concepts like landscape, evening, and nature; and activities like a person getting out of a car or a person skiing. The video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartLabelDetection returns a job identifier (JobId) which you use to get the results of the operation. When label detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the label detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetLabelDetection and pass the job identifier (JobId) from the initial call to StartLabelDetection. Optional Parameters StartLabelDetection has the GENERAL_LABELS Feature applied by default. This feature allows you to provide filtering criteria to the Settings parameter. You can filter with sets of individual labels or with label categories. You can specify inclusive filters, exclusive filters, or a combination of inclusive and exclusive filters. For more information on filtering, see Detecting labels in a video. You can specify MinConfidence to control the confidence threshold for the labels returned. The default is 50.

Starts the asynchronous search for faces in a collection that match the faces of persons detected in a stored video. The video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartFaceSearch returns a job identifier (JobId) which you use to get the search results once the search has completed. When searching is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the search results, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceSearch and pass the job identifier (JobId) from the initial call to StartFaceSearch. For more information, see Searching stored videos for faces.

Starts the asynchronous search for faces in a collection that match the faces of persons detected in a stored video. The video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartFaceSearch returns a job identifier (JobId) which you use to get the search results once the search has completed. When searching is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the search results, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceSearch and pass the job identifier (JobId) from the initial call to StartFaceSearch. For more information, see Searching stored videos for faces.

Starts asynchronous detection of faces in a stored video. Amazon Rekognition Video can detect faces in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartFaceDetection returns a job identifier (JobId) that you use to get the results of the operation. When face detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the face detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceDetection and pass the job identifier (JobId) from the initial call to StartFaceDetection. For more information, see Detecting faces in a stored video in the Amazon Rekognition Developer Guide.

Starts asynchronous detection of faces in a stored video. Amazon Rekognition Video can detect faces in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartFaceDetection returns a job identifier (JobId) that you use to get the results of the operation. When face detection is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the face detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceDetection and pass the job identifier (JobId) from the initial call to StartFaceDetection. For more information, see Detecting faces in a stored video in the Amazon Rekognition Developer Guide.

Starts asynchronous detection of inappropriate, unwanted, or offensive content in a stored video. For a list of moderation labels in Amazon Rekognition, see Using the image and video moderation APIs. Amazon Rekognition Video can moderate content in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartContentModeration returns a job identifier (JobId) which you use to get the results of the analysis. When content analysis is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the content analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetContentModeration and pass the job identifier (JobId) from the initial call to StartContentModeration. For more information, see Moderating content in the Amazon Rekognition Developer Guide.

Starts asynchronous detection of inappropriate, unwanted, or offensive content in a stored video. For a list of moderation labels in Amazon Rekognition, see Using the image and video moderation APIs. Amazon Rekognition Video can moderate content in a video stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartContentModeration returns a job identifier (JobId) which you use to get the results of the analysis. When content analysis is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the content analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetContentModeration and pass the job identifier (JobId) from the initial call to StartContentModeration. For more information, see Moderating content in the Amazon Rekognition Developer Guide.

Starts asynchronous recognition of celebrities in a stored video. Amazon Rekognition Video can detect celebrities in a video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartCelebrityRecognition returns a job identifier (JobId) which you use to get the results of the analysis. When celebrity recognition analysis is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the celebrity recognition analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetCelebrityRecognition and pass the job identifier (JobId) from the initial call to StartCelebrityRecognition. For more information, see Recognizing celebrities in the Amazon Rekognition Developer Guide.

Starts asynchronous recognition of celebrities in a stored video. Amazon Rekognition Video can detect celebrities in a video must be stored in an Amazon S3 bucket. Use Video to specify the bucket name and the filename of the video. StartCelebrityRecognition returns a job identifier (JobId) which you use to get the results of the analysis. When celebrity recognition analysis is finished, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic that you specify in NotificationChannel. To get the results of the celebrity recognition analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetCelebrityRecognition and pass the job identifier (JobId) from the initial call to StartCelebrityRecognition. For more information, see Recognizing celebrities in the Amazon Rekognition Developer Guide.

Searches for UserIDs within a collection based on a FaceId or UserId. This API can be used to find the closest UserID (with a highest similarity) to associate a face. The request must be provided with either FaceId or UserId. The operation returns an array of UserID that match the FaceId or UserId, ordered by similarity score with the highest similarity first.

Searches for UserIDs within a collection based on a FaceId or UserId. This API can be used to find the closest UserID (with a highest similarity) to associate a face. The request must be provided with either FaceId or UserId. The operation returns an array of UserID that match the FaceId or UserId, ordered by similarity score with the highest similarity first.

Searches for UserIDs using a supplied image. It first detects the largest face in the image, and then searches a specified collection for matching UserIDs. The operation returns an array of UserIDs that match the face in the supplied image, ordered by similarity score with the highest similarity first. It also returns a bounding box for the face found in the input image. Information about faces detected in the supplied image, but not used for the search, is returned in an array of UnsearchedFace objects. If no valid face is detected in the image, the response will contain an empty UserMatches list and no SearchedFace object.

Searches for UserIDs using a supplied image. It first detects the largest face in the image, and then searches a specified collection for matching UserIDs. The operation returns an array of UserIDs that match the face in the supplied image, ordered by similarity score with the highest similarity first. It also returns a bounding box for the face found in the input image. Information about faces detected in the supplied image, but not used for the search, is returned in an array of UnsearchedFace objects. If no valid face is detected in the image, the response will contain an empty UserMatches list and no SearchedFace object.

For a given input face ID, searches for matching faces in the collection the face belongs to. You get a face ID when you add a face to the collection using the IndexFaces operation. The operation compares the features of the input face with faces in the specified collection. You can also search faces without indexing faces by using the SearchFacesByImage operation. The operation response returns an array of faces that match, ordered by similarity score with the highest similarity first. More specifically, it is an array of metadata for each face match that is found. Along with the metadata, the response also includes a confidence value for each face match, indicating the confidence that the specific face matches the input face. For an example, see Searching for a face using its face ID in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:SearchFaces action.

For a given input face ID, searches for matching faces in the collection the face belongs to. You get a face ID when you add a face to the collection using the IndexFaces operation. The operation compares the features of the input face with faces in the specified collection. You can also search faces without indexing faces by using the SearchFacesByImage operation. The operation response returns an array of faces that match, ordered by similarity score with the highest similarity first. More specifically, it is an array of metadata for each face match that is found. Along with the metadata, the response also includes a confidence value for each face match, indicating the confidence that the specific face matches the input face. For an example, see Searching for a face using its face ID in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:SearchFaces action.

For a given input image, first detects the largest face in the image, and then searches the specified collection for matching faces. The operation compares the features of the input face with faces in the specified collection. To search for all faces in an input image, you might first call the IndexFaces operation, and then use the face IDs returned in subsequent calls to the SearchFaces operation. You can also call the DetectFaces operation and use the bounding boxes in the response to make face crops, which then you can pass in to the SearchFacesByImage operation. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. The response returns an array of faces that match, ordered by similarity score with the highest similarity first. More specifically, it is an array of metadata for each face match found. Along with the metadata, the response also includes a similarity indicating how similar the face is to the input face. In the response, the operation also returns the bounding box (and a confidence level that the bounding box contains a face) of the face that Amazon Rekognition used for the input image. If no faces are detected in the input image, SearchFacesByImage returns an InvalidParameterException error. For an example, Searching for a Face Using an Image in the Amazon Rekognition Developer Guide. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. Use QualityFilter to set the quality bar for filtering by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. The default value is NONE. To use quality filtering, you need a collection associated with version 3 of the face model or higher. To get the version of the face model associated with a collection, call DescribeCollection. This operation requires permissions to perform the rekognition:SearchFacesByImage action.

For a given input image, first detects the largest face in the image, and then searches the specified collection for matching faces. The operation compares the features of the input face with faces in the specified collection. To search for all faces in an input image, you might first call the IndexFaces operation, and then use the face IDs returned in subsequent calls to the SearchFaces operation. You can also call the DetectFaces operation and use the bounding boxes in the response to make face crops, which then you can pass in to the SearchFacesByImage operation. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. The response returns an array of faces that match, ordered by similarity score with the highest similarity first. More specifically, it is an array of metadata for each face match found. Along with the metadata, the response also includes a similarity indicating how similar the face is to the input face. In the response, the operation also returns the bounding box (and a confidence level that the bounding box contains a face) of the face that Amazon Rekognition used for the input image. If no faces are detected in the input image, SearchFacesByImage returns an InvalidParameterException error. For an example, Searching for a Face Using an Image in the Amazon Rekognition Developer Guide. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. Use QualityFilter to set the quality bar for filtering by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. The default value is NONE. To use quality filtering, you need a collection associated with version 3 of the face model or higher. To get the version of the face model associated with a collection, call DescribeCollection. This operation requires permissions to perform the rekognition:SearchFacesByImage action.

Returns an array of celebrities recognized in the input image. For more information, see Recognizing celebrities in the Amazon Rekognition Developer Guide. RecognizeCelebrities returns the 64 largest faces in the image. It lists the recognized celebrities in the CelebrityFaces array and any unrecognized faces in the UnrecognizedFaces array. RecognizeCelebrities doesn't return celebrities whose faces aren't among the largest 64 faces in the image. For each celebrity recognized, RecognizeCelebrities returns a Celebrity object. The Celebrity object contains the celebrity name, ID, URL links to additional information, match confidence, and a ComparedFace object that you can use to locate the celebrity's face on the image. Amazon Rekognition doesn't retain information about which images a celebrity has been recognized in. Your application must store this information and use the Celebrity ID property as a unique identifier for the celebrity. If you don't store the celebrity name or additional information URLs returned by RecognizeCelebrities, you will need the ID to identify the celebrity in a call to the GetCelebrityInfo operation. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. For an example, see Recognizing celebrities in an image in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:RecognizeCelebrities operation.

Returns an array of celebrities recognized in the input image. For more information, see Recognizing celebrities in the Amazon Rekognition Developer Guide. RecognizeCelebrities returns the 64 largest faces in the image. It lists the recognized celebrities in the CelebrityFaces array and any unrecognized faces in the UnrecognizedFaces array. RecognizeCelebrities doesn't return celebrities whose faces aren't among the largest 64 faces in the image. For each celebrity recognized, RecognizeCelebrities returns a Celebrity object. The Celebrity object contains the celebrity name, ID, URL links to additional information, match confidence, and a ComparedFace object that you can use to locate the celebrity's face on the image. Amazon Rekognition doesn't retain information about which images a celebrity has been recognized in. Your application must store this information and use the Celebrity ID property as a unique identifier for the celebrity. If you don't store the celebrity name or additional information URLs returned by RecognizeCelebrities, you will need the ID to identify the celebrity in a call to the GetCelebrityInfo operation. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. For an example, see Recognizing celebrities in an image in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:RecognizeCelebrities operation.

This operation applies only to Amazon Rekognition Custom Labels. Attaches a project policy to a Amazon Rekognition Custom Labels project in a trusting AWS account. A project policy specifies that a trusted AWS account can copy a model version from a trusting AWS account to a project in the trusted AWS account. To copy a model version you use the CopyProjectVersion operation. Only applies to Custom Labels projects. For more information about the format of a project policy document, see Attaching a project policy (SDK) in the Amazon Rekognition Custom Labels Developer Guide. The response from PutProjectPolicy is a revision ID for the project policy. You can attach multiple project policies to a project. You can also update an existing project policy by specifying the policy revision ID of the existing policy. To remove a project policy from a project, call DeleteProjectPolicy. To get a list of project policies attached to a project, call ListProjectPolicies. You copy a model version by calling CopyProjectVersion. This operation requires permissions to perform the rekognition:PutProjectPolicy action.

This operation applies only to Amazon Rekognition Custom Labels. Attaches a project policy to a Amazon Rekognition Custom Labels project in a trusting AWS account. A project policy specifies that a trusted AWS account can copy a model version from a trusting AWS account to a project in the trusted AWS account. To copy a model version you use the CopyProjectVersion operation. Only applies to Custom Labels projects. For more information about the format of a project policy document, see Attaching a project policy (SDK) in the Amazon Rekognition Custom Labels Developer Guide. The response from PutProjectPolicy is a revision ID for the project policy. You can attach multiple project policies to a project. You can also update an existing project policy by specifying the policy revision ID of the existing policy. To remove a project policy from a project, call DeleteProjectPolicy. To get a list of project policies attached to a project, call ListProjectPolicies. You copy a model version by calling CopyProjectVersion. This operation requires permissions to perform the rekognition:PutProjectPolicy action.

Returns metadata of the User such as UserID in the specified collection. Anonymous User (to reserve faces without any identity) is not returned as part of this request. The results are sorted by system generated primary key ID. If the response is truncated, NextToken is returned in the response that can be used in the subsequent request to retrieve the next set of identities.

Returns metadata of the User such as UserID in the specified collection. Anonymous User (to reserve faces without any identity) is not returned as part of this request. The results are sorted by system generated primary key ID. If the response is truncated, NextToken is returned in the response that can be used in the subsequent request to retrieve the next set of identities.

Returns a list of tags in an Amazon Rekognition collection, stream processor, or Custom Labels model. This operation requires permissions to perform the rekognition:ListTagsForResource action.

Returns a list of tags in an Amazon Rekognition collection, stream processor, or Custom Labels model. This operation requires permissions to perform the rekognition:ListTagsForResource action.

Gets a list of stream processors that you have created with CreateStreamProcessor.

Gets a list of stream processors that you have created with CreateStreamProcessor.

This operation applies only to Amazon Rekognition Custom Labels. Gets a list of the project policies attached to a project. To attach a project policy to a project, call PutProjectPolicy. To remove a project policy from a project, call DeleteProjectPolicy. This operation requires permissions to perform the rekognition:ListProjectPolicies action.

This operation applies only to Amazon Rekognition Custom Labels. Gets a list of the project policies attached to a project. To attach a project policy to a project, call PutProjectPolicy. To remove a project policy from a project, call DeleteProjectPolicy. This operation requires permissions to perform the rekognition:ListProjectPolicies action.

Returns a list of media analysis jobs. Results are sorted by CreationTimestamp in descending order.

Returns a list of media analysis jobs. Results are sorted by CreationTimestamp in descending order.

Returns metadata for faces in the specified collection. This metadata includes information such as the bounding box coordinates, the confidence (that the bounding box contains a face), and face ID. For an example, see Listing Faces in a Collection in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:ListFaces action.

Returns metadata for faces in the specified collection. This metadata includes information such as the bounding box coordinates, the confidence (that the bounding box contains a face), and face ID. For an example, see Listing Faces in a Collection in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:ListFaces action.

This operation applies only to Amazon Rekognition Custom Labels. Lists the labels in a dataset. Amazon Rekognition Custom Labels uses labels to describe images. For more information, see Labeling images. Lists the labels in a dataset. Amazon Rekognition Custom Labels uses labels to describe images. For more information, see Labeling images in the Amazon Rekognition Custom Labels Developer Guide.

This operation applies only to Amazon Rekognition Custom Labels. Lists the labels in a dataset. Amazon Rekognition Custom Labels uses labels to describe images. For more information, see Labeling images. Lists the labels in a dataset. Amazon Rekognition Custom Labels uses labels to describe images. For more information, see Labeling images in the Amazon Rekognition Custom Labels Developer Guide.

This operation applies only to Amazon Rekognition Custom Labels. Lists the entries (images) within a dataset. An entry is a JSON Line that contains the information for a single image, including the image location, assigned labels, and object location bounding boxes. For more information, see Creating a manifest file. JSON Lines in the response include information about non-terminal errors found in the dataset. Non terminal errors are reported in errors lists within each JSON Line. The same information is reported in the training and testing validation result manifests that Amazon Rekognition Custom Labels creates during model training. You can filter the response in variety of ways, such as choosing which labels to return and returning JSON Lines created after a specific date. This operation requires permissions to perform the rekognition:ListDatasetEntries action.

This operation applies only to Amazon Rekognition Custom Labels. Lists the entries (images) within a dataset. An entry is a JSON Line that contains the information for a single image, including the image location, assigned labels, and object location bounding boxes. For more information, see Creating a manifest file. JSON Lines in the response include information about non-terminal errors found in the dataset. Non terminal errors are reported in errors lists within each JSON Line. The same information is reported in the training and testing validation result manifests that Amazon Rekognition Custom Labels creates during model training. You can filter the response in variety of ways, such as choosing which labels to return and returning JSON Lines created after a specific date. This operation requires permissions to perform the rekognition:ListDatasetEntries action.

Returns list of collection IDs in your account. If the result is truncated, the response also provides a NextToken that you can use in the subsequent request to fetch the next set of collection IDs. For an example, see Listing collections in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:ListCollections action.

Returns list of collection IDs in your account. If the result is truncated, the response also provides a NextToken that you can use in the subsequent request to fetch the next set of collection IDs. For an example, see Listing collections in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:ListCollections action.

Detects faces in the input image and adds them to the specified collection. Amazon Rekognition doesn't save the actual faces that are detected. Instead, the underlying detection algorithm first detects the faces in the input image. For each face, the algorithm extracts facial features into a feature vector, and stores it in the backend database. Amazon Rekognition uses feature vectors when it performs face match and search operations using the SearchFaces and SearchFacesByImage operations. For more information, see Adding faces to a collection in the Amazon Rekognition Developer Guide. To get the number of faces in a collection, call DescribeCollection. If you're using version 1.0 of the face detection model, IndexFaces indexes the 15 largest faces in the input image. Later versions of the face detection model index the 100 largest faces in the input image. If you're using version 4 or later of the face model, image orientation information is not returned in the OrientationCorrection field. To determine which version of the model you're using, call DescribeCollection and supply the collection ID. You can also get the model version from the value of FaceModelVersion in the response from IndexFaces For more information, see Model Versioning in the Amazon Rekognition Developer Guide. If you provide the optional ExternalImageId for the input image you provided, Amazon Rekognition associates this ID with all faces that it detects. When you call the ListFaces operation, the response returns the external ID. You can use this external image ID to create a client-side index to associate the faces with each image. You can then use the index to find all faces in an image. You can specify the maximum number of faces to index with the MaxFaces input parameter. This is useful when you want to index the largest faces in an image and don't want to index smaller faces, such as those belonging to people standing in the background. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. By default, IndexFaces chooses the quality bar that's used to filter faces. You can also explicitly choose the quality bar. Use QualityFilter, to set the quality bar by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. To use quality filtering, you need a collection associated with version 3 of the face model or higher. To get the version of the face model associated with a collection, call DescribeCollection. Information about faces detected in an image, but not indexed, is returned in an array of UnindexedFace objects, UnindexedFaces. Faces aren't indexed for reasons such as: The number of faces detected exceeds the value of the MaxFaces request parameter. The face is too small compared to the image dimensions. The face is too blurry. The image is too dark. The face has an extreme pose. The face doesn’t have enough detail to be suitable for face search. In response, the IndexFaces operation returns an array of metadata for all detected faces, FaceRecords. This includes: The bounding box, BoundingBox, of the detected face. A confidence value, Confidence, which indicates the confidence that the bounding box contains a face. A face ID, FaceId, assigned by the service for each face that's detected and stored. An image ID, ImageId, assigned by the service for the input image. If you request ALL or specific facial attributes (e.g., FACE_OCCLUDED) by using the detectionAttributes parameter, Amazon Rekognition returns detailed facial attributes, such as facial landmarks (for example, location of eye and mouth), facial occlusion, and other facial attributes. If you provide the same image, specify the same collection, and use the same external ID in the IndexFaces operation, Amazon Rekognition doesn't save duplicate face metadata. The input image is passed either as base64-encoded image bytes, or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes isn't supported. The image must be formatted as a PNG or JPEG file. This operation requires permissions to perform the rekognition:IndexFaces action.

Detects faces in the input image and adds them to the specified collection. Amazon Rekognition doesn't save the actual faces that are detected. Instead, the underlying detection algorithm first detects the faces in the input image. For each face, the algorithm extracts facial features into a feature vector, and stores it in the backend database. Amazon Rekognition uses feature vectors when it performs face match and search operations using the SearchFaces and SearchFacesByImage operations. For more information, see Adding faces to a collection in the Amazon Rekognition Developer Guide. To get the number of faces in a collection, call DescribeCollection. If you're using version 1.0 of the face detection model, IndexFaces indexes the 15 largest faces in the input image. Later versions of the face detection model index the 100 largest faces in the input image. If you're using version 4 or later of the face model, image orientation information is not returned in the OrientationCorrection field. To determine which version of the model you're using, call DescribeCollection and supply the collection ID. You can also get the model version from the value of FaceModelVersion in the response from IndexFaces For more information, see Model Versioning in the Amazon Rekognition Developer Guide. If you provide the optional ExternalImageId for the input image you provided, Amazon Rekognition associates this ID with all faces that it detects. When you call the ListFaces operation, the response returns the external ID. You can use this external image ID to create a client-side index to associate the faces with each image. You can then use the index to find all faces in an image. You can specify the maximum number of faces to index with the MaxFaces input parameter. This is useful when you want to index the largest faces in an image and don't want to index smaller faces, such as those belonging to people standing in the background. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. By default, IndexFaces chooses the quality bar that's used to filter faces. You can also explicitly choose the quality bar. Use QualityFilter, to set the quality bar by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. To use quality filtering, you need a collection associated with version 3 of the face model or higher. To get the version of the face model associated with a collection, call DescribeCollection. Information about faces detected in an image, but not indexed, is returned in an array of UnindexedFace objects, UnindexedFaces. Faces aren't indexed for reasons such as: The number of faces detected exceeds the value of the MaxFaces request parameter. The face is too small compared to the image dimensions. The face is too blurry. The image is too dark. The face has an extreme pose. The face doesn’t have enough detail to be suitable for face search. In response, the IndexFaces operation returns an array of metadata for all detected faces, FaceRecords. This includes: The bounding box, BoundingBox, of the detected face. A confidence value, Confidence, which indicates the confidence that the bounding box contains a face. A face ID, FaceId, assigned by the service for each face that's detected and stored. An image ID, ImageId, assigned by the service for the input image. If you request ALL or specific facial attributes (e.g., FACE_OCCLUDED) by using the detectionAttributes parameter, Amazon Rekognition returns detailed facial attributes, such as facial landmarks (for example, location of eye and mouth), facial occlusion, and other facial attributes. If you provide the same image, specify the same collection, and use the same external ID in the IndexFaces operation, Amazon Rekognition doesn't save duplicate face metadata. The input image is passed either as base64-encoded image bytes, or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes isn't supported. The image must be formatted as a PNG or JPEG file. This operation requires permissions to perform the rekognition:IndexFaces action.

Gets the text detection results of a Amazon Rekognition Video analysis started by StartTextDetection. Text detection with Amazon Rekognition Video is an asynchronous operation. You start text detection by calling StartTextDetection which returns a job identifier (JobId) When the text detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartTextDetection. To get the results of the text detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetTextDetection and pass the job identifier (JobId) from the initial call of StartLabelDetection. GetTextDetection returns an array of detected text (TextDetections) sorted by the time the text was detected, up to 100 words per frame of video. Each element of the array includes the detected text, the precentage confidence in the acuracy of the detected text, the time the text was detected, bounding box information for where the text was located, and unique identifiers for words and their lines. Use MaxResults parameter to limit the number of text detections returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetTextDetection and populate the NextToken request parameter with the token value returned from the previous call to GetTextDetection.

Gets the text detection results of a Amazon Rekognition Video analysis started by StartTextDetection. Text detection with Amazon Rekognition Video is an asynchronous operation. You start text detection by calling StartTextDetection which returns a job identifier (JobId) When the text detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartTextDetection. To get the results of the text detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetTextDetection and pass the job identifier (JobId) from the initial call of StartLabelDetection. GetTextDetection returns an array of detected text (TextDetections) sorted by the time the text was detected, up to 100 words per frame of video. Each element of the array includes the detected text, the precentage confidence in the acuracy of the detected text, the time the text was detected, bounding box information for where the text was located, and unique identifiers for words and their lines. Use MaxResults parameter to limit the number of text detections returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetTextDetection and populate the NextToken request parameter with the token value returned from the previous call to GetTextDetection.

Gets the segment detection results of a Amazon Rekognition Video analysis started by StartSegmentDetection. Segment detection with Amazon Rekognition Video is an asynchronous operation. You start segment detection by calling StartSegmentDetection which returns a job identifier (JobId). When the segment detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartSegmentDetection. To get the results of the segment detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetSegmentDetection and pass the job identifier (JobId) from the initial call of StartSegmentDetection. GetSegmentDetection returns detected segments in an array (Segments) of SegmentDetection objects. Segments is sorted by the segment types specified in the SegmentTypes input parameter of StartSegmentDetection. Each element of the array includes the detected segment, the precentage confidence in the acuracy of the detected segment, the type of the segment, and the frame in which the segment was detected. Use SelectedSegmentTypes to find out the type of segment detection requested in the call to StartSegmentDetection. Use the MaxResults parameter to limit the number of segment detections returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetSegmentDetection and populate the NextToken request parameter with the token value returned from the previous call to GetSegmentDetection. For more information, see Detecting video segments in stored video in the Amazon Rekognition Developer Guide.

Gets the segment detection results of a Amazon Rekognition Video analysis started by StartSegmentDetection. Segment detection with Amazon Rekognition Video is an asynchronous operation. You start segment detection by calling StartSegmentDetection which returns a job identifier (JobId). When the segment detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartSegmentDetection. To get the results of the segment detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. if so, call GetSegmentDetection and pass the job identifier (JobId) from the initial call of StartSegmentDetection. GetSegmentDetection returns detected segments in an array (Segments) of SegmentDetection objects. Segments is sorted by the segment types specified in the SegmentTypes input parameter of StartSegmentDetection. Each element of the array includes the detected segment, the precentage confidence in the acuracy of the detected segment, the type of the segment, and the frame in which the segment was detected. Use SelectedSegmentTypes to find out the type of segment detection requested in the call to StartSegmentDetection. Use the MaxResults parameter to limit the number of segment detections returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetSegmentDetection and populate the NextToken request parameter with the token value returned from the previous call to GetSegmentDetection. For more information, see Detecting video segments in stored video in the Amazon Rekognition Developer Guide.

End of support notice: On October 31, 2025, AWS will discontinue support for Amazon Rekognition People Pathing. After October 31, 2025, you will no longer be able to use the Rekognition People Pathing capability. For more information, visit this blog post. Gets the path tracking results of a Amazon Rekognition Video analysis started by StartPersonTracking. The person path tracking operation is started by a call to StartPersonTracking which returns a job identifier (JobId). When the operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartPersonTracking. To get the results of the person path tracking operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetPersonTracking and pass the job identifier (JobId) from the initial call to StartPersonTracking. GetPersonTracking returns an array, Persons, of tracked persons and the time(s) their paths were tracked in the video. GetPersonTracking only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the array is sorted by the time(s) a person's path is tracked in the video. You can sort by tracked persons by specifying INDEX for the SortBy input parameter. Use the MaxResults parameter to limit the number of items returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetPersonTracking and populate the NextToken request parameter with the token value returned from the previous call to GetPersonTracking.

End of support notice: On October 31, 2025, AWS will discontinue support for Amazon Rekognition People Pathing. After October 31, 2025, you will no longer be able to use the Rekognition People Pathing capability. For more information, visit this blog post. Gets the path tracking results of a Amazon Rekognition Video analysis started by StartPersonTracking. The person path tracking operation is started by a call to StartPersonTracking which returns a job identifier (JobId). When the operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartPersonTracking. To get the results of the person path tracking operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetPersonTracking and pass the job identifier (JobId) from the initial call to StartPersonTracking. GetPersonTracking returns an array, Persons, of tracked persons and the time(s) their paths were tracked in the video. GetPersonTracking only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the array is sorted by the time(s) a person's path is tracked in the video. You can sort by tracked persons by specifying INDEX for the SortBy input parameter. Use the MaxResults parameter to limit the number of items returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetPersonTracking and populate the NextToken request parameter with the token value returned from the previous call to GetPersonTracking.

Retrieves the results for a given media analysis job. Takes a JobId returned by StartMediaAnalysisJob.

Retrieves the results for a given media analysis job. Takes a JobId returned by StartMediaAnalysisJob.

Gets the label detection results of a Amazon Rekognition Video analysis started by StartLabelDetection. The label detection operation is started by a call to StartLabelDetection which returns a job identifier (JobId). When the label detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartlabelDetection. To get the results of the label detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetLabelDetection and pass the job identifier (JobId) from the initial call to StartLabelDetection. GetLabelDetection returns an array of detected labels (Labels) sorted by the time the labels were detected. You can also sort by the label name by specifying NAME for the SortBy input parameter. If there is no NAME specified, the default sort is by timestamp. You can select how results are aggregated by using the AggregateBy input parameter. The default aggregation method is TIMESTAMPS. You can also aggregate by SEGMENTS, which aggregates all instances of labels detected in a given segment. The returned Labels array may include the following attributes: Name - The name of the detected label. Confidence - The level of confidence in the label assigned to a detected object. Parents - The ancestor labels for a detected label. GetLabelDetection returns a hierarchical taxonomy of detected labels. For example, a detected car might be assigned the label car. The label car has two parent labels: Vehicle (its parent) and Transportation (its grandparent). The response includes the all ancestors for a label, where every ancestor is a unique label. In the previous example, Car, Vehicle, and Transportation are returned as unique labels in the response. Aliases - Possible Aliases for the label. Categories - The label categories that the detected label belongs to. BoundingBox — Bounding boxes are described for all instances of detected common object labels, returned in an array of Instance objects. An Instance object contains a BoundingBox object, describing the location of the label on the input image. It also includes the confidence for the accuracy of the detected bounding box. Timestamp - Time, in milliseconds from the start of the video, that the label was detected. For aggregation by SEGMENTS, the StartTimestampMillis, EndTimestampMillis, and DurationMillis structures are what define a segment. Although the “Timestamp” structure is still returned with each label, its value is set to be the same as StartTimestampMillis. Timestamp and Bounding box information are returned for detected Instances, only if aggregation is done by TIMESTAMPS. If aggregating by SEGMENTS, information about detected instances isn’t returned. The version of the label model used for the detection is also returned. Note DominantColors isn't returned for Instances, although it is shown as part of the response in the sample seen below. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetlabelDetection and populate the NextToken request parameter with the token value returned from the previous call to GetLabelDetection. If you are retrieving results while using the Amazon Simple Notification Service, note that you will receive an "ERROR" notification if the job encounters an issue.

Gets the label detection results of a Amazon Rekognition Video analysis started by StartLabelDetection. The label detection operation is started by a call to StartLabelDetection which returns a job identifier (JobId). When the label detection operation finishes, Amazon Rekognition publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartlabelDetection. To get the results of the label detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetLabelDetection and pass the job identifier (JobId) from the initial call to StartLabelDetection. GetLabelDetection returns an array of detected labels (Labels) sorted by the time the labels were detected. You can also sort by the label name by specifying NAME for the SortBy input parameter. If there is no NAME specified, the default sort is by timestamp. You can select how results are aggregated by using the AggregateBy input parameter. The default aggregation method is TIMESTAMPS. You can also aggregate by SEGMENTS, which aggregates all instances of labels detected in a given segment. The returned Labels array may include the following attributes: Name - The name of the detected label. Confidence - The level of confidence in the label assigned to a detected object. Parents - The ancestor labels for a detected label. GetLabelDetection returns a hierarchical taxonomy of detected labels. For example, a detected car might be assigned the label car. The label car has two parent labels: Vehicle (its parent) and Transportation (its grandparent). The response includes the all ancestors for a label, where every ancestor is a unique label. In the previous example, Car, Vehicle, and Transportation are returned as unique labels in the response. Aliases - Possible Aliases for the label. Categories - The label categories that the detected label belongs to. BoundingBox — Bounding boxes are described for all instances of detected common object labels, returned in an array of Instance objects. An Instance object contains a BoundingBox object, describing the location of the label on the input image. It also includes the confidence for the accuracy of the detected bounding box. Timestamp - Time, in milliseconds from the start of the video, that the label was detected. For aggregation by SEGMENTS, the StartTimestampMillis, EndTimestampMillis, and DurationMillis structures are what define a segment. Although the “Timestamp” structure is still returned with each label, its value is set to be the same as StartTimestampMillis. Timestamp and Bounding box information are returned for detected Instances, only if aggregation is done by TIMESTAMPS. If aggregating by SEGMENTS, information about detected instances isn’t returned. The version of the label model used for the detection is also returned. Note DominantColors isn't returned for Instances, although it is shown as part of the response in the sample seen below. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetlabelDetection and populate the NextToken request parameter with the token value returned from the previous call to GetLabelDetection. If you are retrieving results while using the Amazon Simple Notification Service, note that you will receive an "ERROR" notification if the job encounters an issue.

Gets the face search results for Amazon Rekognition Video face search started by StartFaceSearch. The search returns faces in a collection that match the faces of persons detected in a video. It also includes the time(s) that faces are matched in the video. Face search in a video is an asynchronous operation. You start face search by calling to StartFaceSearch which returns a job identifier (JobId). When the search operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartFaceSearch. To get the search results, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceSearch and pass the job identifier (JobId) from the initial call to StartFaceSearch. For more information, see Searching Faces in a Collection in the Amazon Rekognition Developer Guide. The search results are retured in an array, Persons, of PersonMatch objects. EachPersonMatch element contains details about the matching faces in the input collection, person information (facial attributes, bounding boxes, and person identifer) for the matched person, and the time the person was matched in the video. GetFaceSearch only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the Persons array is sorted by the time, in milliseconds from the start of the video, persons are matched. You can also sort by persons by specifying INDEX for the SORTBY input parameter.

Gets the face search results for Amazon Rekognition Video face search started by StartFaceSearch. The search returns faces in a collection that match the faces of persons detected in a video. It also includes the time(s) that faces are matched in the video. Face search in a video is an asynchronous operation. You start face search by calling to StartFaceSearch which returns a job identifier (JobId). When the search operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartFaceSearch. To get the search results, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceSearch and pass the job identifier (JobId) from the initial call to StartFaceSearch. For more information, see Searching Faces in a Collection in the Amazon Rekognition Developer Guide. The search results are retured in an array, Persons, of PersonMatch objects. EachPersonMatch element contains details about the matching faces in the input collection, person information (facial attributes, bounding boxes, and person identifer) for the matched person, and the time the person was matched in the video. GetFaceSearch only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the Persons array is sorted by the time, in milliseconds from the start of the video, persons are matched. You can also sort by persons by specifying INDEX for the SORTBY input parameter.

Retrieves the results of a specific Face Liveness session. It requires the sessionId as input, which was created using CreateFaceLivenessSession. Returns the corresponding Face Liveness confidence score, a reference image that includes a face bounding box, and audit images that also contain face bounding boxes. The Face Liveness confidence score ranges from 0 to 100. The number of audit images returned by GetFaceLivenessSessionResults is defined by the AuditImagesLimit paramater when calling CreateFaceLivenessSession. Reference images are always returned when possible.

Retrieves the results of a specific Face Liveness session. It requires the sessionId as input, which was created using CreateFaceLivenessSession. Returns the corresponding Face Liveness confidence score, a reference image that includes a face bounding box, and audit images that also contain face bounding boxes. The Face Liveness confidence score ranges from 0 to 100. The number of audit images returned by GetFaceLivenessSessionResults is defined by the AuditImagesLimit paramater when calling CreateFaceLivenessSession. Reference images are always returned when possible.

Gets face detection results for a Amazon Rekognition Video analysis started by StartFaceDetection. Face detection with Amazon Rekognition Video is an asynchronous operation. You start face detection by calling StartFaceDetection which returns a job identifier (JobId). When the face detection operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartFaceDetection. To get the results of the face detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceDetection and pass the job identifier (JobId) from the initial call to StartFaceDetection. GetFaceDetection returns an array of detected faces (Faces) sorted by the time the faces were detected. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetFaceDetection and populate the NextToken request parameter with the token value returned from the previous call to GetFaceDetection. Note that for the GetFaceDetection operation, the returned values for FaceOccluded and EyeDirection will always be "null".

Gets face detection results for a Amazon Rekognition Video analysis started by StartFaceDetection. Face detection with Amazon Rekognition Video is an asynchronous operation. You start face detection by calling StartFaceDetection which returns a job identifier (JobId). When the face detection operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartFaceDetection. To get the results of the face detection operation, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetFaceDetection and pass the job identifier (JobId) from the initial call to StartFaceDetection. GetFaceDetection returns an array of detected faces (Faces) sorted by the time the faces were detected. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetFaceDetection and populate the NextToken request parameter with the token value returned from the previous call to GetFaceDetection. Note that for the GetFaceDetection operation, the returned values for FaceOccluded and EyeDirection will always be "null".

Gets the inappropriate, unwanted, or offensive content analysis results for a Amazon Rekognition Video analysis started by StartContentModeration. For a list of moderation labels in Amazon Rekognition, see Using the image and video moderation APIs. Amazon Rekognition Video inappropriate or offensive content detection in a stored video is an asynchronous operation. You start analysis by calling StartContentModeration which returns a job identifier (JobId). When analysis finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartContentModeration. To get the results of the content analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetContentModeration and pass the job identifier (JobId) from the initial call to StartContentModeration. For more information, see Working with Stored Videos in the Amazon Rekognition Devlopers Guide. GetContentModeration returns detected inappropriate, unwanted, or offensive content moderation labels, and the time they are detected, in an array, ModerationLabels, of ContentModerationDetection objects. By default, the moderated labels are returned sorted by time, in milliseconds from the start of the video. You can also sort them by moderated label by specifying NAME for the SortBy input parameter. Since video analysis can return a large number of results, use the MaxResults parameter to limit the number of labels returned in a single call to GetContentModeration. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetContentModeration and populate the NextToken request parameter with the value of NextToken returned from the previous call to GetContentModeration. For more information, see moderating content in the Amazon Rekognition Developer Guide.

Gets the inappropriate, unwanted, or offensive content analysis results for a Amazon Rekognition Video analysis started by StartContentModeration. For a list of moderation labels in Amazon Rekognition, see Using the image and video moderation APIs. Amazon Rekognition Video inappropriate or offensive content detection in a stored video is an asynchronous operation. You start analysis by calling StartContentModeration which returns a job identifier (JobId). When analysis finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartContentModeration. To get the results of the content analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetContentModeration and pass the job identifier (JobId) from the initial call to StartContentModeration. For more information, see Working with Stored Videos in the Amazon Rekognition Devlopers Guide. GetContentModeration returns detected inappropriate, unwanted, or offensive content moderation labels, and the time they are detected, in an array, ModerationLabels, of ContentModerationDetection objects. By default, the moderated labels are returned sorted by time, in milliseconds from the start of the video. You can also sort them by moderated label by specifying NAME for the SortBy input parameter. Since video analysis can return a large number of results, use the MaxResults parameter to limit the number of labels returned in a single call to GetContentModeration. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetContentModeration and populate the NextToken request parameter with the value of NextToken returned from the previous call to GetContentModeration. For more information, see moderating content in the Amazon Rekognition Developer Guide.

Gets the celebrity recognition results for a Amazon Rekognition Video analysis started by StartCelebrityRecognition. Celebrity recognition in a video is an asynchronous operation. Analysis is started by a call to StartCelebrityRecognition which returns a job identifier (JobId). When the celebrity recognition operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartCelebrityRecognition. To get the results of the celebrity recognition analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetCelebrityDetection and pass the job identifier (JobId) from the initial call to StartCelebrityDetection. For more information, see Working With Stored Videos in the Amazon Rekognition Developer Guide. GetCelebrityRecognition returns detected celebrities and the time(s) they are detected in an array (Celebrities) of CelebrityRecognition objects. Each CelebrityRecognition contains information about the celebrity in a CelebrityDetail object and the time, Timestamp, the celebrity was detected. This CelebrityDetail object stores information about the detected celebrity's face attributes, a face bounding box, known gender, the celebrity's name, and a confidence estimate. GetCelebrityRecognition only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The BoundingBox field only applies to the detected face instance. The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the Celebrities array is sorted by time (milliseconds from the start of the video). You can also sort the array by celebrity by specifying the value ID in the SortBy input parameter. The CelebrityDetail object includes the celebrity identifer and additional information urls. If you don't store the additional information urls, you can get them later by calling GetCelebrityInfo with the celebrity identifer. No information is returned for faces not recognized as celebrities. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetCelebrityDetection and populate the NextToken request parameter with the token value returned from the previous call to GetCelebrityRecognition.

Gets the celebrity recognition results for a Amazon Rekognition Video analysis started by StartCelebrityRecognition. Celebrity recognition in a video is an asynchronous operation. Analysis is started by a call to StartCelebrityRecognition which returns a job identifier (JobId). When the celebrity recognition operation finishes, Amazon Rekognition Video publishes a completion status to the Amazon Simple Notification Service topic registered in the initial call to StartCelebrityRecognition. To get the results of the celebrity recognition analysis, first check that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetCelebrityDetection and pass the job identifier (JobId) from the initial call to StartCelebrityDetection. For more information, see Working With Stored Videos in the Amazon Rekognition Developer Guide. GetCelebrityRecognition returns detected celebrities and the time(s) they are detected in an array (Celebrities) of CelebrityRecognition objects. Each CelebrityRecognition contains information about the celebrity in a CelebrityDetail object and the time, Timestamp, the celebrity was detected. This CelebrityDetail object stores information about the detected celebrity's face attributes, a face bounding box, known gender, the celebrity's name, and a confidence estimate. GetCelebrityRecognition only returns the default facial attributes (BoundingBox, Confidence, Landmarks, Pose, and Quality). The BoundingBox field only applies to the detected face instance. The other facial attributes listed in the Face object of the following response syntax are not returned. For more information, see FaceDetail in the Amazon Rekognition Developer Guide. By default, the Celebrities array is sorted by time (milliseconds from the start of the video). You can also sort the array by celebrity by specifying the value ID in the SortBy input parameter. The CelebrityDetail object includes the celebrity identifer and additional information urls. If you don't store the additional information urls, you can get them later by calling GetCelebrityInfo with the celebrity identifer. No information is returned for faces not recognized as celebrities. Use MaxResults parameter to limit the number of labels returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetCelebrityDetection and populate the NextToken request parameter with the token value returned from the previous call to GetCelebrityRecognition.

Gets the name and additional information about a celebrity based on their Amazon Rekognition ID. The additional information is returned as an array of URLs. If there is no additional information about the celebrity, this list is empty. For more information, see Getting information about a celebrity in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:GetCelebrityInfo action.

Gets the name and additional information about a celebrity based on their Amazon Rekognition ID. The additional information is returned as an array of URLs. If there is no additional information about the celebrity, this list is empty. For more information, see Getting information about a celebrity in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:GetCelebrityInfo action.

This operation applies only to Amazon Rekognition Custom Labels. Distributes the entries (images) in a training dataset across the training dataset and the test dataset for a project. DistributeDatasetEntries moves 20% of the training dataset images to the test dataset. An entry is a JSON Line that describes an image. You supply the Amazon Resource Names (ARN) of a project's training dataset and test dataset. The training dataset must contain the images that you want to split. The test dataset must be empty. The datasets must belong to the same project. To create training and test datasets for a project, call CreateDataset. Distributing a dataset takes a while to complete. To check the status call DescribeDataset. The operation is complete when the Status field for the training dataset and the test dataset is UPDATE_COMPLETE. If the dataset split fails, the value of Status is UPDATE_FAILED. This operation requires permissions to perform the rekognition:DistributeDatasetEntries action.

This operation applies only to Amazon Rekognition Custom Labels. Distributes the entries (images) in a training dataset across the training dataset and the test dataset for a project. DistributeDatasetEntries moves 20% of the training dataset images to the test dataset. An entry is a JSON Line that describes an image. You supply the Amazon Resource Names (ARN) of a project's training dataset and test dataset. The training dataset must contain the images that you want to split. The test dataset must be empty. The datasets must belong to the same project. To create training and test datasets for a project, call CreateDataset. Distributing a dataset takes a while to complete. To check the status call DescribeDataset. The operation is complete when the Status field for the training dataset and the test dataset is UPDATE_COMPLETE. If the dataset split fails, the value of Status is UPDATE_FAILED. This operation requires permissions to perform the rekognition:DistributeDatasetEntries action.

Removes the association between a Face supplied in an array of FaceIds and the User. If the User is not present already, then a ResourceNotFound exception is thrown. If successful, an array of faces that are disassociated from the User is returned. If a given face is already disassociated from the given UserID, it will be ignored and not be returned in the response. If a given face is already associated with a different User or not found in the collection it will be returned as part of UnsuccessfulDisassociations. You can remove 1 - 100 face IDs from a user at one time.

Removes the association between a Face supplied in an array of FaceIds and the User. If the User is not present already, then a ResourceNotFound exception is thrown. If successful, an array of faces that are disassociated from the User is returned. If a given face is already disassociated from the given UserID, it will be ignored and not be returned in the response. If a given face is already associated with a different User or not found in the collection it will be returned as part of UnsuccessfulDisassociations. You can remove 1 - 100 face IDs from a user at one time.

Detects text in the input image and converts it into machine-readable text. Pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, you must pass it as a reference to an image in an Amazon S3 bucket. For the AWS CLI, passing image bytes is not supported. The image must be either a .png or .jpeg formatted file. The DetectText operation returns text in an array of TextDetection elements, TextDetections. Each TextDetection element provides information about a single word or line of text that was detected in the image. A word is one or more script characters that are not separated by spaces. DetectText can detect up to 100 words in an image. A line is a string of equally spaced words. A line isn't necessarily a complete sentence. For example, a driver's license number is detected as a line. A line ends when there is no aligned text after it. Also, a line ends when there is a large gap between words, relative to the length of the words. This means, depending on the gap between words, Amazon Rekognition may detect multiple lines in text aligned in the same direction. Periods don't represent the end of a line. If a sentence spans multiple lines, the DetectText operation returns multiple lines. To determine whether a TextDetection element is a line of text or a word, use the TextDetection object Type field. To be detected, text must be within +/- 90 degrees orientation of the horizontal axis. For more information, see Detecting text in the Amazon Rekognition Developer Guide.

Detects text in the input image and converts it into machine-readable text. Pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, you must pass it as a reference to an image in an Amazon S3 bucket. For the AWS CLI, passing image bytes is not supported. The image must be either a .png or .jpeg formatted file. The DetectText operation returns text in an array of TextDetection elements, TextDetections. Each TextDetection element provides information about a single word or line of text that was detected in the image. A word is one or more script characters that are not separated by spaces. DetectText can detect up to 100 words in an image. A line is a string of equally spaced words. A line isn't necessarily a complete sentence. For example, a driver's license number is detected as a line. A line ends when there is no aligned text after it. Also, a line ends when there is a large gap between words, relative to the length of the words. This means, depending on the gap between words, Amazon Rekognition may detect multiple lines in text aligned in the same direction. Periods don't represent the end of a line. If a sentence spans multiple lines, the DetectText operation returns multiple lines. To determine whether a TextDetection element is a line of text or a word, use the TextDetection object Type field. To be detected, text must be within +/- 90 degrees orientation of the horizontal axis. For more information, see Detecting text in the Amazon Rekognition Developer Guide.

Detects Personal Protective Equipment (PPE) worn by people detected in an image. Amazon Rekognition can detect the following types of PPE. Face cover Hand cover Head cover You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. The image must be either a PNG or JPG formatted file. DetectProtectiveEquipment detects PPE worn by up to 15 persons detected in an image. For each person detected in the image the API returns an array of body parts (face, head, left-hand, right-hand). For each body part, an array of detected items of PPE is returned, including an indicator of whether or not the PPE covers the body part. The API returns the confidence it has in each detection (person, PPE, body part and body part coverage). It also returns a bounding box (BoundingBox) for each detected person and each detected item of PPE. You can optionally request a summary of detected PPE items with the SummarizationAttributes input parameter. The summary provides the following information. The persons detected as wearing all of the types of PPE that you specify. The persons detected as not wearing all of the types PPE that you specify. The persons detected where PPE adornment could not be determined. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectProtectiveEquipment action.

Detects Personal Protective Equipment (PPE) worn by people detected in an image. Amazon Rekognition can detect the following types of PPE. Face cover Hand cover Head cover You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. The image must be either a PNG or JPG formatted file. DetectProtectiveEquipment detects PPE worn by up to 15 persons detected in an image. For each person detected in the image the API returns an array of body parts (face, head, left-hand, right-hand). For each body part, an array of detected items of PPE is returned, including an indicator of whether or not the PPE covers the body part. The API returns the confidence it has in each detection (person, PPE, body part and body part coverage). It also returns a bounding box (BoundingBox) for each detected person and each detected item of PPE. You can optionally request a summary of detected PPE items with the SummarizationAttributes input parameter. The summary provides the following information. The persons detected as wearing all of the types of PPE that you specify. The persons detected as not wearing all of the types PPE that you specify. The persons detected where PPE adornment could not be determined. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectProtectiveEquipment action.

Detects unsafe content in a specified JPEG or PNG format image. Use DetectModerationLabels to moderate images depending on your requirements. For example, you might want to filter images that contain nudity, but not images containing suggestive content. To filter images, use the labels returned by DetectModerationLabels to determine which types of content are appropriate. For information about moderation labels, see Detecting Unsafe Content in the Amazon Rekognition Developer Guide. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. You can specify an adapter to use when retrieving label predictions by providing a ProjectVersionArn to the ProjectVersion argument.

Detects unsafe content in a specified JPEG or PNG format image. Use DetectModerationLabels to moderate images depending on your requirements. For example, you might want to filter images that contain nudity, but not images containing suggestive content. To filter images, use the labels returned by DetectModerationLabels to determine which types of content are appropriate. For information about moderation labels, see Detecting Unsafe Content in the Amazon Rekognition Developer Guide. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. You can specify an adapter to use when retrieving label predictions by providing a ProjectVersionArn to the ProjectVersion argument.

Detects instances of real-world entities within an image (JPEG or PNG) provided as input. This includes objects like flower, tree, and table; events like wedding, graduation, and birthday party; and concepts like landscape, evening, and nature. For an example, see Analyzing images stored in an Amazon S3 bucket in the Amazon Rekognition Developer Guide. You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. Optional Parameters You can specify one or both of the GENERAL_LABELS and IMAGE_PROPERTIES feature types when calling the DetectLabels API. Including GENERAL_LABELS will ensure the response includes the labels detected in the input image, while including IMAGE_PROPERTIES will ensure the response includes information about the image quality and color. When using GENERAL_LABELS and/or IMAGE_PROPERTIES you can provide filtering criteria to the Settings parameter. You can filter with sets of individual labels or with label categories. You can specify inclusive filters, exclusive filters, or a combination of inclusive and exclusive filters. For more information on filtering see Detecting Labels in an Image. When getting labels, you can specify MinConfidence to control the confidence threshold for the labels returned. The default is 55%. You can also add the MaxLabels parameter to limit the number of labels returned. The default and upper limit is 1000 labels. These arguments are only valid when supplying GENERAL_LABELS as a feature type. Response Elements For each object, scene, and concept the API returns one or more labels. The API returns the following types of information about labels: Name - The name of the detected label. Confidence - The level of confidence in the label assigned to a detected object. Parents - The ancestor labels for a detected label. DetectLabels returns a hierarchical taxonomy of detected labels. For example, a detected car might be assigned the label car. The label car has two parent labels: Vehicle (its parent) and Transportation (its grandparent). The response includes the all ancestors for a label, where every ancestor is a unique label. In the previous example, Car, Vehicle, and Transportation are returned as unique labels in the response. Aliases - Possible Aliases for the label. Categories - The label categories that the detected label belongs to. BoundingBox — Bounding boxes are described for all instances of detected common object labels, returned in an array of Instance objects. An Instance object contains a BoundingBox object, describing the location of the label on the input image. It also includes the confidence for the accuracy of the detected bounding box. The API returns the following information regarding the image, as part of the ImageProperties structure: Quality - Information about the Sharpness, Brightness, and Contrast of the input image, scored between 0 to 100. Image quality is returned for the entire image, as well as the background and the foreground. Dominant Color - An array of the dominant colors in the image. Foreground - Information about the sharpness, brightness, and dominant colors of the input image’s foreground. Background - Information about the sharpness, brightness, and dominant colors of the input image’s background. The list of returned labels will include at least one label for every detected object, along with information about that label. In the following example, suppose the input image has a lighthouse, the sea, and a rock. The response includes all three labels, one for each object, as well as the confidence in the label: {Name: lighthouse, Confidence: 98.4629} {Name: rock,Confidence: 79.2097} {Name: sea,Confidence: 75.061} The list of labels can include multiple labels for the same object. For example, if the input image shows a flower (for example, a tulip), the operation might return the following three labels. {Name: flower,Confidence: 99.0562} {Name: plant,Confidence: 99.0562} {Name: tulip,Confidence: 99.0562} In this example, the detection algorithm more precisely identifies the flower as a tulip. If the object detected is a person, the operation doesn't provide the same facial details that the DetectFaces operation provides. This is a stateless API operation that doesn't return any data. This operation requires permissions to perform the rekognition:DetectLabels action.

Detects instances of real-world entities within an image (JPEG or PNG) provided as input. This includes objects like flower, tree, and table; events like wedding, graduation, and birthday party; and concepts like landscape, evening, and nature. For an example, see Analyzing images stored in an Amazon S3 bucket in the Amazon Rekognition Developer Guide. You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. Optional Parameters You can specify one or both of the GENERAL_LABELS and IMAGE_PROPERTIES feature types when calling the DetectLabels API. Including GENERAL_LABELS will ensure the response includes the labels detected in the input image, while including IMAGE_PROPERTIES will ensure the response includes information about the image quality and color. When using GENERAL_LABELS and/or IMAGE_PROPERTIES you can provide filtering criteria to the Settings parameter. You can filter with sets of individual labels or with label categories. You can specify inclusive filters, exclusive filters, or a combination of inclusive and exclusive filters. For more information on filtering see Detecting Labels in an Image. When getting labels, you can specify MinConfidence to control the confidence threshold for the labels returned. The default is 55%. You can also add the MaxLabels parameter to limit the number of labels returned. The default and upper limit is 1000 labels. These arguments are only valid when supplying GENERAL_LABELS as a feature type. Response Elements For each object, scene, and concept the API returns one or more labels. The API returns the following types of information about labels: Name - The name of the detected label. Confidence - The level of confidence in the label assigned to a detected object. Parents - The ancestor labels for a detected label. DetectLabels returns a hierarchical taxonomy of detected labels. For example, a detected car might be assigned the label car. The label car has two parent labels: Vehicle (its parent) and Transportation (its grandparent). The response includes the all ancestors for a label, where every ancestor is a unique label. In the previous example, Car, Vehicle, and Transportation are returned as unique labels in the response. Aliases - Possible Aliases for the label. Categories - The label categories that the detected label belongs to. BoundingBox — Bounding boxes are described for all instances of detected common object labels, returned in an array of Instance objects. An Instance object contains a BoundingBox object, describing the location of the label on the input image. It also includes the confidence for the accuracy of the detected bounding box. The API returns the following information regarding the image, as part of the ImageProperties structure: Quality - Information about the Sharpness, Brightness, and Contrast of the input image, scored between 0 to 100. Image quality is returned for the entire image, as well as the background and the foreground. Dominant Color - An array of the dominant colors in the image. Foreground - Information about the sharpness, brightness, and dominant colors of the input image’s foreground. Background - Information about the sharpness, brightness, and dominant colors of the input image’s background. The list of returned labels will include at least one label for every detected object, along with information about that label. In the following example, suppose the input image has a lighthouse, the sea, and a rock. The response includes all three labels, one for each object, as well as the confidence in the label: {Name: lighthouse, Confidence: 98.4629} {Name: rock,Confidence: 79.2097} {Name: sea,Confidence: 75.061} The list of labels can include multiple labels for the same object. For example, if the input image shows a flower (for example, a tulip), the operation might return the following three labels. {Name: flower,Confidence: 99.0562} {Name: plant,Confidence: 99.0562} {Name: tulip,Confidence: 99.0562} In this example, the detection algorithm more precisely identifies the flower as a tulip. If the object detected is a person, the operation doesn't provide the same facial details that the DetectFaces operation provides. This is a stateless API operation that doesn't return any data. This operation requires permissions to perform the rekognition:DetectLabels action.

Detects faces within an image that is provided as input. DetectFaces detects the 100 largest faces in the image. For each face detected, the operation returns face details. These details include a bounding box of the face, a confidence value (that the bounding box contains a face), and a fixed set of attributes such as facial landmarks (for example, coordinates of eye and mouth), pose, presence of facial occlusion, and so on. The face-detection algorithm is most effective on frontal faces. For non-frontal or obscured faces, the algorithm might not detect the faces or might detect faces with lower confidence. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectFaces action.

Detects faces within an image that is provided as input. DetectFaces detects the 100 largest faces in the image. For each face detected, the operation returns face details. These details include a bounding box of the face, a confidence value (that the bounding box contains a face), and a fixed set of attributes such as facial landmarks (for example, coordinates of eye and mouth), pose, presence of facial occlusion, and so on. The face-detection algorithm is most effective on frontal faces. For non-frontal or obscured faces, the algorithm might not detect the faces or might detect faces with lower confidence. You pass the input image either as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectFaces action.

This operation applies only to Amazon Rekognition Custom Labels. Detects custom labels in a supplied image by using an Amazon Rekognition Custom Labels model. You specify which version of a model version to use by using the ProjectVersionArn input parameter. You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. For each object that the model version detects on an image, the API returns a (CustomLabel) object in an array (CustomLabels). Each CustomLabel object provides the label name (Name), the level of confidence that the image contains the object (Confidence), and object location information, if it exists, for the label on the image (Geometry). To filter labels that are returned, specify a value for MinConfidence. DetectCustomLabelsLabels only returns labels with a confidence that's higher than the specified value. The value of MinConfidence maps to the assumed threshold values created during training. For more information, see Assumed threshold in the Amazon Rekognition Custom Labels Developer Guide. Amazon Rekognition Custom Labels metrics expresses an assumed threshold as a floating point value between 0-1. The range of MinConfidence normalizes the threshold value to a percentage value (0-100). Confidence responses from DetectCustomLabels are also returned as a percentage. You can use MinConfidence to change the precision and recall or your model. For more information, see Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide. If you don't specify a value for MinConfidence, DetectCustomLabels returns labels based on the assumed threshold of each label. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectCustomLabels action. For more information, see Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide.

This operation applies only to Amazon Rekognition Custom Labels. Detects custom labels in a supplied image by using an Amazon Rekognition Custom Labels model. You specify which version of a model version to use by using the ProjectVersionArn input parameter. You pass the input image as base64-encoded image bytes or as a reference to an image in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes is not supported. The image must be either a PNG or JPEG formatted file. For each object that the model version detects on an image, the API returns a (CustomLabel) object in an array (CustomLabels). Each CustomLabel object provides the label name (Name), the level of confidence that the image contains the object (Confidence), and object location information, if it exists, for the label on the image (Geometry). To filter labels that are returned, specify a value for MinConfidence. DetectCustomLabelsLabels only returns labels with a confidence that's higher than the specified value. The value of MinConfidence maps to the assumed threshold values created during training. For more information, see Assumed threshold in the Amazon Rekognition Custom Labels Developer Guide. Amazon Rekognition Custom Labels metrics expresses an assumed threshold as a floating point value between 0-1. The range of MinConfidence normalizes the threshold value to a percentage value (0-100). Confidence responses from DetectCustomLabels are also returned as a percentage. You can use MinConfidence to change the precision and recall or your model. For more information, see Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide. If you don't specify a value for MinConfidence, DetectCustomLabels returns labels based on the assumed threshold of each label. This is a stateless API operation. That is, the operation does not persist any data. This operation requires permissions to perform the rekognition:DetectCustomLabels action. For more information, see Analyzing an image in the Amazon Rekognition Custom Labels Developer Guide.

Provides information about a stream processor created by CreateStreamProcessor. You can get information about the input and output streams, the input parameters for the face recognition being performed, and the current status of the stream processor.

Provides information about a stream processor created by CreateStreamProcessor. You can get information about the input and output streams, the input parameters for the face recognition being performed, and the current status of the stream processor.

Gets information about your Rekognition projects. This operation requires permissions to perform the rekognition:DescribeProjects action.

Gets information about your Rekognition projects. This operation requires permissions to perform the rekognition:DescribeProjects action.

Lists and describes the versions of an Amazon Rekognition project. You can specify up to 10 model or adapter versions in ProjectVersionArns. If you don't specify a value, descriptions for all model/adapter versions in the project are returned. This operation requires permissions to perform the rekognition:DescribeProjectVersions action.

Lists and describes the versions of an Amazon Rekognition project. You can specify up to 10 model or adapter versions in ProjectVersionArns. If you don't specify a value, descriptions for all model/adapter versions in the project are returned. This operation requires permissions to perform the rekognition:DescribeProjectVersions action.

This operation applies only to Amazon Rekognition Custom Labels. Describes an Amazon Rekognition Custom Labels dataset. You can get information such as the current status of a dataset and statistics about the images and labels in a dataset. This operation requires permissions to perform the rekognition:DescribeDataset action.

This operation applies only to Amazon Rekognition Custom Labels. Describes an Amazon Rekognition Custom Labels dataset. You can get information such as the current status of a dataset and statistics about the images and labels in a dataset. This operation requires permissions to perform the rekognition:DescribeDataset action.

Describes the specified collection. You can use DescribeCollection to get information, such as the number of faces indexed into a collection and the version of the model used by the collection for face detection. For more information, see Describing a Collection in the Amazon Rekognition Developer Guide.

Describes the specified collection. You can use DescribeCollection to get information, such as the number of faces indexed into a collection and the version of the model used by the collection for face detection. For more information, see Describing a Collection in the Amazon Rekognition Developer Guide.

Deletes the specified UserID within the collection. Faces that are associated with the UserID are disassociated from the UserID before deleting the specified UserID. If the specified Collection or UserID is already deleted or not found, a ResourceNotFoundException will be thrown. If the action is successful with a 200 response, an empty HTTP body is returned.

Deletes the specified UserID within the collection. Faces that are associated with the UserID are disassociated from the UserID before deleting the specified UserID. If the specified Collection or UserID is already deleted or not found, a ResourceNotFoundException will be thrown. If the action is successful with a 200 response, an empty HTTP body is returned.

Deletes the stream processor identified by Name. You assign the value for Name when you create the stream processor with CreateStreamProcessor. You might not be able to use the same name for a stream processor for a few seconds after calling DeleteStreamProcessor.

Deletes the stream processor identified by Name. You assign the value for Name when you create the stream processor with CreateStreamProcessor. You might not be able to use the same name for a stream processor for a few seconds after calling DeleteStreamProcessor.

Deletes a Rekognition project model or project version, like a Amazon Rekognition Custom Labels model or a custom adapter. You can't delete a project version if it is running or if it is training. To check the status of a project version, use the Status field returned from DescribeProjectVersions. To stop a project version call StopProjectVersion. If the project version is training, wait until it finishes. This operation requires permissions to perform the rekognition:DeleteProjectVersion action.

Deletes a Rekognition project model or project version, like a Amazon Rekognition Custom Labels model or a custom adapter. You can't delete a project version if it is running or if it is training. To check the status of a project version, use the Status field returned from DescribeProjectVersions. To stop a project version call StopProjectVersion. If the project version is training, wait until it finishes. This operation requires permissions to perform the rekognition:DeleteProjectVersion action.

Deletes a Amazon Rekognition project. To delete a project you must first delete all models or adapters associated with the project. To delete a model or adapter, see DeleteProjectVersion. DeleteProject is an asynchronous operation. To check if the project is deleted, call DescribeProjects. The project is deleted when the project no longer appears in the response. Be aware that deleting a given project will also delete any ProjectPolicies associated with that project. This operation requires permissions to perform the rekognition:DeleteProject action.

Deletes a Amazon Rekognition project. To delete a project you must first delete all models or adapters associated with the project. To delete a model or adapter, see DeleteProjectVersion. DeleteProject is an asynchronous operation. To check if the project is deleted, call DescribeProjects. The project is deleted when the project no longer appears in the response. Be aware that deleting a given project will also delete any ProjectPolicies associated with that project. This operation requires permissions to perform the rekognition:DeleteProject action.

This operation applies only to Amazon Rekognition Custom Labels. Deletes an existing project policy. To get a list of project policies attached to a project, call ListProjectPolicies. To attach a project policy to a project, call PutProjectPolicy. This operation requires permissions to perform the rekognition:DeleteProjectPolicy action.

This operation applies only to Amazon Rekognition Custom Labels. Deletes an existing project policy. To get a list of project policies attached to a project, call ListProjectPolicies. To attach a project policy to a project, call PutProjectPolicy. This operation requires permissions to perform the rekognition:DeleteProjectPolicy action.

Deletes faces from a collection. You specify a collection ID and an array of face IDs to remove from the collection. This operation requires permissions to perform the rekognition:DeleteFaces action.

Deletes faces from a collection. You specify a collection ID and an array of face IDs to remove from the collection. This operation requires permissions to perform the rekognition:DeleteFaces action.

This operation applies only to Amazon Rekognition Custom Labels. Deletes an existing Amazon Rekognition Custom Labels dataset. Deleting a dataset might take while. Use DescribeDataset to check the current status. The dataset is still deleting if the value of Status is DELETE_IN_PROGRESS. If you try to access the dataset after it is deleted, you get a ResourceNotFoundException exception. You can't delete a dataset while it is creating (Status = CREATE_IN_PROGRESS) or if the dataset is updating (Status = UPDATE_IN_PROGRESS). This operation requires permissions to perform the rekognition:DeleteDataset action.

This operation applies only to Amazon Rekognition Custom Labels. Deletes an existing Amazon Rekognition Custom Labels dataset. Deleting a dataset might take while. Use DescribeDataset to check the current status. The dataset is still deleting if the value of Status is DELETE_IN_PROGRESS. If you try to access the dataset after it is deleted, you get a ResourceNotFoundException exception. You can't delete a dataset while it is creating (Status = CREATE_IN_PROGRESS) or if the dataset is updating (Status = UPDATE_IN_PROGRESS). This operation requires permissions to perform the rekognition:DeleteDataset action.

Deletes the specified collection. Note that this operation removes all faces in the collection. For an example, see Deleting a collection. This operation requires permissions to perform the rekognition:DeleteCollection action.

Deletes the specified collection. Note that this operation removes all faces in the collection. For an example, see Deleting a collection. This operation requires permissions to perform the rekognition:DeleteCollection action.

Creates a new User within a collection specified by CollectionId. Takes UserId as a parameter, which is a user provided ID which should be unique within the collection. The provided UserId will alias the system generated UUID to make the UserId more user friendly. Uses a ClientToken, an idempotency token that ensures a call to CreateUser completes only once. If the value is not supplied, the AWS SDK generates an idempotency token for the requests. This prevents retries after a network error results from making multiple CreateUser calls.

Creates a new User within a collection specified by CollectionId. Takes UserId as a parameter, which is a user provided ID which should be unique within the collection. The provided UserId will alias the system generated UUID to make the UserId more user friendly. Uses a ClientToken, an idempotency token that ensures a call to CreateUser completes only once. If the value is not supplied, the AWS SDK generates an idempotency token for the requests. This prevents retries after a network error results from making multiple CreateUser calls.

Creates an Amazon Rekognition stream processor that you can use to detect and recognize faces or to detect labels in a streaming video. Amazon Rekognition Video is a consumer of live video from Amazon Kinesis Video Streams. There are two different settings for stream processors in Amazon Rekognition: detecting faces and detecting labels. If you are creating a stream processor for detecting faces, you provide as input a Kinesis video stream (Input) and a Kinesis data stream (Output) stream for receiving the output. You must use the FaceSearch option in Settings, specifying the collection that contains the faces you want to recognize. After you have finished analyzing a streaming video, use StopStreamProcessor to stop processing. If you are creating a stream processor to detect labels, you provide as input a Kinesis video stream (Input), Amazon S3 bucket information (Output), and an Amazon SNS topic ARN (NotificationChannel). You can also provide a KMS key ID to encrypt the data sent to your Amazon S3 bucket. You specify what you want to detect by using the ConnectedHome option in settings, and selecting one of the following: PERSON, PET, PACKAGE, ALL You can also specify where in the frame you want Amazon Rekognition to monitor with RegionsOfInterest. When you run the StartStreamProcessor operation on a label detection stream processor, you input start and stop information to determine the length of the processing time. Use Name to assign an identifier for the stream processor. You use Name to manage the stream processor. For example, you can start processing the source video by calling StartStreamProcessor with the Name field. This operation requires permissions to perform the rekognition:CreateStreamProcessor action. If you want to tag your stream processor, you also require permission to perform the rekognition:TagResource operation.

Creates an Amazon Rekognition stream processor that you can use to detect and recognize faces or to detect labels in a streaming video. Amazon Rekognition Video is a consumer of live video from Amazon Kinesis Video Streams. There are two different settings for stream processors in Amazon Rekognition: detecting faces and detecting labels. If you are creating a stream processor for detecting faces, you provide as input a Kinesis video stream (Input) and a Kinesis data stream (Output) stream for receiving the output. You must use the FaceSearch option in Settings, specifying the collection that contains the faces you want to recognize. After you have finished analyzing a streaming video, use StopStreamProcessor to stop processing. If you are creating a stream processor to detect labels, you provide as input a Kinesis video stream (Input), Amazon S3 bucket information (Output), and an Amazon SNS topic ARN (NotificationChannel). You can also provide a KMS key ID to encrypt the data sent to your Amazon S3 bucket. You specify what you want to detect by using the ConnectedHome option in settings, and selecting one of the following: PERSON, PET, PACKAGE, ALL You can also specify where in the frame you want Amazon Rekognition to monitor with RegionsOfInterest. When you run the StartStreamProcessor operation on a label detection stream processor, you input start and stop information to determine the length of the processing time. Use Name to assign an identifier for the stream processor. You use Name to manage the stream processor. For example, you can start processing the source video by calling StartStreamProcessor with the Name field. This operation requires permissions to perform the rekognition:CreateStreamProcessor action. If you want to tag your stream processor, you also require permission to perform the rekognition:TagResource operation.

Creates a new version of Amazon Rekognition project (like a Custom Labels model or a custom adapter) and begins training. Models and adapters are managed as part of a Rekognition project. The response from CreateProjectVersion is an Amazon Resource Name (ARN) for the project version. The FeatureConfig operation argument allows you to configure specific model or adapter settings. You can provide a description to the project version by using the VersionDescription argment. Training can take a while to complete. You can get the current status by calling DescribeProjectVersions. Training completed successfully if the value of the Status field is TRAINING_COMPLETED. Once training has successfully completed, call DescribeProjectVersions to get the training results and evaluate the model. This operation requires permissions to perform the rekognition:CreateProjectVersion action. The following applies only to projects with Amazon Rekognition Custom Labels as the chosen feature: You can train a model in a project that doesn't have associated datasets by specifying manifest files in the TrainingData and TestingData fields. If you open the console after training a model with manifest files, Amazon Rekognition Custom Labels creates the datasets for you using the most recent manifest files. You can no longer train a model version for the project by specifying manifest files. Instead of training with a project without associated datasets, we recommend that you use the manifest files to create training and test datasets for the project.

Creates a new version of Amazon Rekognition project (like a Custom Labels model or a custom adapter) and begins training. Models and adapters are managed as part of a Rekognition project. The response from CreateProjectVersion is an Amazon Resource Name (ARN) for the project version. The FeatureConfig operation argument allows you to configure specific model or adapter settings. You can provide a description to the project version by using the VersionDescription argment. Training can take a while to complete. You can get the current status by calling DescribeProjectVersions. Training completed successfully if the value of the Status field is TRAINING_COMPLETED. Once training has successfully completed, call DescribeProjectVersions to get the training results and evaluate the model. This operation requires permissions to perform the rekognition:CreateProjectVersion action. The following applies only to projects with Amazon Rekognition Custom Labels as the chosen feature: You can train a model in a project that doesn't have associated datasets by specifying manifest files in the TrainingData and TestingData fields. If you open the console after training a model with manifest files, Amazon Rekognition Custom Labels creates the datasets for you using the most recent manifest files. You can no longer train a model version for the project by specifying manifest files. Instead of training with a project without associated datasets, we recommend that you use the manifest files to create training and test datasets for the project.

Creates a new Amazon Rekognition project. A project is a group of resources (datasets, model versions) that you use to create and manage a Amazon Rekognition Custom Labels Model or custom adapter. You can specify a feature to create the project with, if no feature is specified then Custom Labels is used by default. For adapters, you can also choose whether or not to have the project auto update by using the AutoUpdate argument. This operation requires permissions to perform the rekognition:CreateProject action.

Creates a new Amazon Rekognition project. A project is a group of resources (datasets, model versions) that you use to create and manage a Amazon Rekognition Custom Labels Model or custom adapter. You can specify a feature to create the project with, if no feature is specified then Custom Labels is used by default. For adapters, you can also choose whether or not to have the project auto update by using the AutoUpdate argument. This operation requires permissions to perform the rekognition:CreateProject action.

This API operation initiates a Face Liveness session. It returns a SessionId, which you can use to start streaming Face Liveness video and get the results for a Face Liveness session. You can use the OutputConfig option in the Settings parameter to provide an Amazon S3 bucket location. The Amazon S3 bucket stores reference images and audit images. If no Amazon S3 bucket is defined, raw bytes are sent instead. You can use AuditImagesLimit to limit the number of audit images returned when GetFaceLivenessSessionResults is called. This number is between 0 and 4. By default, it is set to 0. The limit is best effort and based on the duration of the selfie-video.

This API operation initiates a Face Liveness session. It returns a SessionId, which you can use to start streaming Face Liveness video and get the results for a Face Liveness session. You can use the OutputConfig option in the Settings parameter to provide an Amazon S3 bucket location. The Amazon S3 bucket stores reference images and audit images. If no Amazon S3 bucket is defined, raw bytes are sent instead. You can use AuditImagesLimit to limit the number of audit images returned when GetFaceLivenessSessionResults is called. This number is between 0 and 4. By default, it is set to 0. The limit is best effort and based on the duration of the selfie-video.

This operation applies only to Amazon Rekognition Custom Labels. Creates a new Amazon Rekognition Custom Labels dataset. You can create a dataset by using an Amazon Sagemaker format manifest file or by copying an existing Amazon Rekognition Custom Labels dataset. To create a training dataset for a project, specify TRAIN for the value of DatasetType. To create the test dataset for a project, specify TEST for the value of DatasetType. The response from CreateDataset is the Amazon Resource Name (ARN) for the dataset. Creating a dataset takes a while to complete. Use DescribeDataset to check the current status. The dataset created successfully if the value of Status is CREATE_COMPLETE. To check if any non-terminal errors occurred, call ListDatasetEntries and check for the presence of errors lists in the JSON Lines. Dataset creation fails if a terminal error occurs (Status = CREATE_FAILED). Currently, you can't access the terminal error information. For more information, see Creating dataset in the Amazon Rekognition Custom Labels Developer Guide. This operation requires permissions to perform the rekognition:CreateDataset action. If you want to copy an existing dataset, you also require permission to perform the rekognition:ListDatasetEntries action.

This operation applies only to Amazon Rekognition Custom Labels. Creates a new Amazon Rekognition Custom Labels dataset. You can create a dataset by using an Amazon Sagemaker format manifest file or by copying an existing Amazon Rekognition Custom Labels dataset. To create a training dataset for a project, specify TRAIN for the value of DatasetType. To create the test dataset for a project, specify TEST for the value of DatasetType. The response from CreateDataset is the Amazon Resource Name (ARN) for the dataset. Creating a dataset takes a while to complete. Use DescribeDataset to check the current status. The dataset created successfully if the value of Status is CREATE_COMPLETE. To check if any non-terminal errors occurred, call ListDatasetEntries and check for the presence of errors lists in the JSON Lines. Dataset creation fails if a terminal error occurs (Status = CREATE_FAILED). Currently, you can't access the terminal error information. For more information, see Creating dataset in the Amazon Rekognition Custom Labels Developer Guide. This operation requires permissions to perform the rekognition:CreateDataset action. If you want to copy an existing dataset, you also require permission to perform the rekognition:ListDatasetEntries action.

Creates a collection in an AWS Region. You can add faces to the collection using the IndexFaces operation. For example, you might create collections, one for each of your application users. A user can then index faces using the IndexFaces operation and persist results in a specific collection. Then, a user can search the collection for faces in the user-specific container. When you create a collection, it is associated with the latest version of the face model version. Collection names are case-sensitive. This operation requires permissions to perform the rekognition:CreateCollection action. If you want to tag your collection, you also require permission to perform the rekognition:TagResource operation.

Creates a collection in an AWS Region. You can add faces to the collection using the IndexFaces operation. For example, you might create collections, one for each of your application users. A user can then index faces using the IndexFaces operation and persist results in a specific collection. Then, a user can search the collection for faces in the user-specific container. When you create a collection, it is associated with the latest version of the face model version. Collection names are case-sensitive. This operation requires permissions to perform the rekognition:CreateCollection action. If you want to tag your collection, you also require permission to perform the rekognition:TagResource operation.

This operation applies only to Amazon Rekognition Custom Labels. Copies a version of an Amazon Rekognition Custom Labels model from a source project to a destination project. The source and destination projects can be in different AWS accounts but must be in the same AWS Region. You can't copy a model to another AWS service. To copy a model version to a different AWS account, you need to create a resource-based policy known as a project policy. You attach the project policy to the source project by calling PutProjectPolicy. The project policy gives permission to copy the model version from a trusting AWS account to a trusted account. For more information creating and attaching a project policy, see Attaching a project policy (SDK) in the Amazon Rekognition Custom Labels Developer Guide. If you are copying a model version to a project in the same AWS account, you don't need to create a project policy. Copying project versions is supported only for Custom Labels models. To copy a model, the destination project, source project, and source model version must already exist. Copying a model version takes a while to complete. To get the current status, call DescribeProjectVersions and check the value of Status in the ProjectVersionDescription object. The copy operation has finished when the value of Status is COPYING_COMPLETED. This operation requires permissions to perform the rekognition:CopyProjectVersion action.

This operation applies only to Amazon Rekognition Custom Labels. Copies a version of an Amazon Rekognition Custom Labels model from a source project to a destination project. The source and destination projects can be in different AWS accounts but must be in the same AWS Region. You can't copy a model to another AWS service. To copy a model version to a different AWS account, you need to create a resource-based policy known as a project policy. You attach the project policy to the source project by calling PutProjectPolicy. The project policy gives permission to copy the model version from a trusting AWS account to a trusted account. For more information creating and attaching a project policy, see Attaching a project policy (SDK) in the Amazon Rekognition Custom Labels Developer Guide. If you are copying a model version to a project in the same AWS account, you don't need to create a project policy. Copying project versions is supported only for Custom Labels models. To copy a model, the destination project, source project, and source model version must already exist. Copying a model version takes a while to complete. To get the current status, call DescribeProjectVersions and check the value of Status in the ProjectVersionDescription object. The copy operation has finished when the value of Status is COPYING_COMPLETED. This operation requires permissions to perform the rekognition:CopyProjectVersion action.

Compares a face in the source input image with each of the 100 largest faces detected in the target input image. If the source image contains multiple faces, the service detects the largest face and compares it with each face detected in the target image. CompareFaces uses machine learning algorithms, which are probabilistic. A false negative is an incorrect prediction that a face in the target image has a low similarity confidence score when compared to the face in the source image. To reduce the probability of false negatives, we recommend that you compare the target image against multiple source images. If you plan to use CompareFaces to make a decision that impacts an individual's rights, privacy, or access to services, we recommend that you pass the result to a human for review and further validation before taking action. You pass the input and target images either as base64-encoded image bytes or as references to images in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes isn't supported. The image must be formatted as a PNG or JPEG file. In response, the operation returns an array of face matches ordered by similarity score in descending order. For each face match, the response provides a bounding box of the face, facial landmarks, pose details (pitch, roll, and yaw), quality (brightness and sharpness), and confidence value (indicating the level of confidence that the bounding box contains a face). The response also provides a similarity score, which indicates how closely the faces match. By default, only faces with a similarity score of greater than or equal to 80% are returned in the response. You can change this value by specifying the SimilarityThreshold parameter. CompareFaces also returns an array of faces that don't match the source image. For each face, it returns a bounding box, confidence value, landmarks, pose details, and quality. The response also returns information about the face in the source image, including the bounding box of the face and confidence value. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. Use QualityFilter to set the quality bar by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. The default value is NONE. If the image doesn't contain Exif metadata, CompareFaces returns orientation information for the source and target images. Use these values to display the images with the correct image orientation. If no faces are detected in the source or target images, CompareFaces returns an InvalidParameterException error. This is a stateless API operation. That is, data returned by this operation doesn't persist. For an example, see Comparing Faces in Images in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:CompareFaces action.

Compares a face in the source input image with each of the 100 largest faces detected in the target input image. If the source image contains multiple faces, the service detects the largest face and compares it with each face detected in the target image. CompareFaces uses machine learning algorithms, which are probabilistic. A false negative is an incorrect prediction that a face in the target image has a low similarity confidence score when compared to the face in the source image. To reduce the probability of false negatives, we recommend that you compare the target image against multiple source images. If you plan to use CompareFaces to make a decision that impacts an individual's rights, privacy, or access to services, we recommend that you pass the result to a human for review and further validation before taking action. You pass the input and target images either as base64-encoded image bytes or as references to images in an Amazon S3 bucket. If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes isn't supported. The image must be formatted as a PNG or JPEG file. In response, the operation returns an array of face matches ordered by similarity score in descending order. For each face match, the response provides a bounding box of the face, facial landmarks, pose details (pitch, roll, and yaw), quality (brightness and sharpness), and confidence value (indicating the level of confidence that the bounding box contains a face). The response also provides a similarity score, which indicates how closely the faces match. By default, only faces with a similarity score of greater than or equal to 80% are returned in the response. You can change this value by specifying the SimilarityThreshold parameter. CompareFaces also returns an array of faces that don't match the source image. For each face, it returns a bounding box, confidence value, landmarks, pose details, and quality. The response also returns information about the face in the source image, including the bounding box of the face and confidence value. The QualityFilter input parameter allows you to filter out detected faces that don’t meet a required quality bar. The quality bar is based on a variety of common use cases. Use QualityFilter to set the quality bar by specifying LOW, MEDIUM, or HIGH. If you do not want to filter detected faces, specify NONE. The default value is NONE. If the image doesn't contain Exif metadata, CompareFaces returns orientation information for the source and target images. Use these values to display the images with the correct image orientation. If no faces are detected in the source or target images, CompareFaces returns an InvalidParameterException error. This is a stateless API operation. That is, data returned by this operation doesn't persist. For an example, see Comparing Faces in Images in the Amazon Rekognition Developer Guide. This operation requires permissions to perform the rekognition:CompareFaces action.

Associates one or more faces with an existing UserID. Takes an array of FaceIds. Each FaceId that are present in the FaceIds list is associated with the provided UserID. The number of FaceIds that can be used as input in a single request is limited to 100. Note that the total number of faces that can be associated with a single UserID is also limited to 100. Once a UserID has 100 faces associated with it, no additional faces can be added. If more API calls are made after the limit is reached, a ServiceQuotaExceededException will result. The UserMatchThreshold parameter specifies the minimum user match confidence required for the face to be associated with a UserID that has at least one FaceID already associated. This ensures that the FaceIds are associated with the right UserID. The value ranges from 0-100 and default value is 75. If successful, an array of AssociatedFace objects containing the associated FaceIds is returned. If a given face is already associated with the given UserID, it will be ignored and will not be returned in the response. If a given face is already associated to a different UserID, isn't found in the collection, doesn’t meet the UserMatchThreshold, or there are already 100 faces associated with the UserID, it will be returned as part of an array of UnsuccessfulFaceAssociations. The UserStatus reflects the status of an operation which updates a UserID representation with a list of given faces. The UserStatus can be: ACTIVE - All associations or disassociations of FaceID(s) for a UserID are complete. CREATED - A UserID has been created, but has no FaceID(s) associated with it. UPDATING - A UserID is being updated and there are current associations or disassociations of FaceID(s) taking place.

Associates one or more faces with an existing UserID. Takes an array of FaceIds. Each FaceId that are present in the FaceIds list is associated with the provided UserID. The number of FaceIds that can be used as input in a single request is limited to 100. Note that the total number of faces that can be associated with a single UserID is also limited to 100. Once a UserID has 100 faces associated with it, no additional faces can be added. If more API calls are made after the limit is reached, a ServiceQuotaExceededException will result. The UserMatchThreshold parameter specifies the minimum user match confidence required for the face to be associated with a UserID that has at least one FaceID already associated. This ensures that the FaceIds are associated with the right UserID. The value ranges from 0-100 and default value is 75. If successful, an array of AssociatedFace objects containing the associated FaceIds is returned. If a given face is already associated with the given UserID, it will be ignored and will not be returned in the response. If a given face is already associated to a different UserID, isn't found in the collection, doesn’t meet the UserMatchThreshold, or there are already 100 faces associated with the UserID, it will be returned as part of an array of UnsuccessfulFaceAssociations. The UserStatus reflects the status of an operation which updates a UserID representation with a list of given faces. The UserStatus can be: ACTIVE - All associations or disassociations of FaceID(s) for a UserID are complete. CREATED - A UserID has been created, but has no FaceID(s) associated with it. UPDATING - A UserID is being updated and there are current associations or disassociations of FaceID(s) taking place.