Amazon Rekognition
Amazon Rekognition is a cloud-based computer vision service that makes it easy to add image and video analysis to your applications, providing capabilities such as object and scene detection, facial analysis, face comparison, celebrity recognition, text detection, content moderation, custom labels, face liveness detection, and streaming video analysis using deep learning technology.
APIs
Amazon Rekognition
Amazon Rekognition provides image and video analysis APIs for label detection, facial analysis, face comparison, celebrity recognition, text detection, content moderation, custo...
Collections
Amazon Rekognition
POSTMANAmazon Rekognition
OPENArazzo Workflows
Amazon Rekognition Celebrity Scene Context
Recognize celebrities in an image and then label the same image for scene context.
ARAZZOAmazon Rekognition Custom Labels and Moderate
Run a Custom Labels model on an image and then screen the same image for unsafe content.
ARAZZOAmazon Rekognition Detect then Compare Faces
Confirm a face exists in the source image, then compare it against every face in a target image.
ARAZZOAmazon Rekognition Enroll and Search a Face
Create a face collection, index a face into it, then search the collection by a query image.
ARAZZOAmazon Rekognition Face Liveness Session
Create a Face Liveness session, then poll for its results until a terminal status is reached.
ARAZZOAmazon Rekognition Label and Moderate an Image
Detect general labels in an image and then screen the same image for unsafe content.
ARAZZOAmazon Rekognition Quality Gated Enrollment
Detect a face and check its quality, then index it into a collection only when a face is present.
ARAZZOAmazon Rekognition Reuse or Create Collection then Enroll
List collections, branch to create the collection only if missing, then index a face into it.
ARAZZOAmazon Rekognition Text and Moderation Screen
Extract text from an image and then screen the same image for unsafe content.
ARAZZOAmazon Rekognition Verify a Face Against a Collection
Detect a face in an image to confirm a single subject, then search a collection to verify identity.
ARAZZOAmazon Rekognition Video Label Detection Job
Start an asynchronous video label detection job, poll until it succeeds, then read the results.
ARAZZOGraphQL
Amazon Rekognition GraphQL Schema
This directory contains a conceptual GraphQL schema for the Amazon Rekognition API. The schema is derived from the Amazon Rekognition REST API and its public documentation at ht...
GRAPHQLPricing Plans
Rate Limits
FinOps
Amazon Rekognition Finops
FINOPSFeatures
Detect thousands of objects, scenes, and concepts in images and videos with high confidence scores using deep learning.
Detect and analyze faces with attributes including age range, emotions, gender, and facial landmarks.
Compare faces across images to determine if they are the same person with a similarity score.
Create searchable face collections to index and search millions of faces in near real-time.
Identify thousands of celebrities in images and videos across categories like sports, entertainment, and politics.
Detect and extract printed and handwritten text from images and videos in multiple languages.
Detect explicit, inappropriate, or violent content in images and videos for automated content moderation.
Build and train custom image classifiers using your own labeled images for domain-specific object detection.
Detect personal protective equipment such as face covers, hand covers, and head covers on persons in images.
Verify that a user is physically present during identity verification to prevent spoofing attacks.
Track and follow identified people across frames in stored video footage.
Identify technical cues and segments such as black frames, end credits, and color bars in video content.
Analyze live streaming video in real-time using Amazon Kinesis Video Streams integration.
Evaluate image quality attributes including sharpness, brightness, contrast, and dominant colors.
Use Cases
Verify user identities by comparing selfies to ID documents or previously stored face images for onboarding and authentication.
Automatically moderate user-generated content on platforms to detect and filter explicit or inappropriate imagery.
Build searchable image and video archives by automatically tagging media with detected labels, faces, and text.
Monitor camera feeds to detect whether workers are wearing required personal protective equipment in industrial settings.
Prevent identity fraud during digital onboarding by using face liveness detection to confirm real users.
Analyze in-store camera feeds to track customer behavior, measure foot traffic, and optimize product placement.
Search video archives for persons of interest by comparing faces against a known collection.
Automatically tag celebrities in photos and videos for media companies to improve content discoverability.
Train custom classifiers to detect proprietary products, logos, brand assets, or industry-specific objects.