Amazon Rekognition
#198 · Level 3 · Modern architecture · AI and machine learningConcept
Section titled “Concept”Amazon Rekognition looks at images and videos and can recognize objects, faces, text and inappropriate content. Its models are already trained, so you can use it through an API without knowing ML. Images or videos are usually kept in S3 and given to Rekognition to analyze. It can compare two faces, and it can also search for a match in a stored collection of faces. There is also a way to train it with your own images to recognize your own special things (Custom Labels).
Simple example
Section titled “Simple example”An attentive guard who looks at a picture and tells you everything that is in it.
Why it’s needed
Section titled “Why it’s needed”Having people look at millions of images is impossible.
Where it’s used
Section titled “Where it’s used”Tagging images, blocking inappropriate images, identity verification.
Azure equivalent
Section titled “Azure equivalent”Azure AI Vision
Learn more
Section titled “Learn more”For details and the latest information, see the official documentation: