Skip to content
This repository has been archived by the owner on Sep 14, 2023. It is now read-only.

Latest commit

 

History

History
102 lines (65 loc) · 5.48 KB

README.md

File metadata and controls

102 lines (65 loc) · 5.48 KB
Raw

Identity Verification with Amazon Rekognition

Warning

This repository is archived and no longer maintained. It may no longer work, or refer to third-party libraries with known problems. We suggest AWS Samples - Rekognition Identity Verification as an alternative to this sample.

This sample, built using AWS Amplify, is meant to showcase recommended flows when using Amazon Rekognition for Identity Verification. There are two key components in this sample:

  • Frontend: A React/NextJS web application for exercising the key flows
  • Backend: An GraphQL based backend built using AWS AppSync

Flows

Users interested in implemented Identity Verification with Amazon Rekognition should consider the following flows.

New user registration

  1. Check face image quality via the DetectFaces API.
  2. Use SearchFacesByImage API against the collection(s) to check for any duplicate registration.
  3. Index the face image using IndexFaces API and use the ExternalImageID (Social Security number or a similar unique ID) parameter to associate the face embeddings with the ExternalImageID.
  4. Store the face image in the S3 bucket along with the user metadata (face-id returned from the IndexFaces API, SSN and S3 URL) in DynamoDB. The SSN or a unique person identifier can be used as a key to lookup S3 URL and the face-id.

New user registration w/ Id card

  1. Check face image quality via the DetectFaces API.
  2. Check face image quality of face on Id card via the DetectFaces API.
  3. Use CompareFaces API to ensure that the face on the supplied Id card and the face on the selfie match.
  4. Use SearchFacesByImage API against the collection(s) to check for any duplicate registration.
  5. Index the face image using IndexFaces API and use the ExternalImageID (Social Security number or a similar unique ID) parameter to associate the face embeddings with the ExternalImageID.
  6. Store the face image in the S3 bucket along with the user metadata (face-id returned from the IndexFaces API, SSN and S3 URL) in DynamoDB. The SSN or a unique person identifier can be used as a key to lookup S3 URL and the face-id.

Existing user login

  1. Check face image quality via the DetectFaces API.
  2. Search against the collection with SearchFacesbyImage API. If there is a face match, then return the use the faceId to return additional data about the user by cross-referencing against profile data in DynamoDB.

Installing and configuring AWS Amplify

Warning

This repository is archived and no longer maintained. It may no longer work, or refer to third-party libraries with known problems. We suggest AWS Samples - Rekognition Identity Verification as an alternative to this sample.

  1. Install npm

First you'll need to install npm, if you don't already have it installed. Please see npm docs for information on installing npm.

  1. Install Amplify

Next, if you don't have the AWS Amplify CLI installed, please follow the instructions at AWS Amplify - Installation.

Deploying sample app using the CLI

Warning

This repository is archived and no longer maintained. It may no longer work, or refer to third-party libraries with known problems. We suggest AWS Samples - Rekognition Identity Verification as an alternative to this sample.

  1. Clone the git repo on your local machine
git clone https://github.com/aws-samples/amazon-rekognition-id-verification-sample-with-amplify
  1. Switch to the cloned directory and run npm update
cd amazon-rekognition-id-verification-sample-with-amplify
npm update --legacy-peer-deps

The --legacy-peer-deps flag is required by a package used in the sample.

  1. Initialize Amplify
amplify init
  1. Deploy the backend
amplify push
  1. Deploy the frontend
amplify add hosting
amplify publish

For more details on hosting/publishing, please see Add hosting to your app. The Amplify Getting Started page has detailed instructions on working with Amplify.

Getting started with the sample app

  1. Create a user in the Cognito User Pool attached to the app.
  2. Accessing the web endpoint. Amplify allows you

Security

See CONTRIBUTING for more information.

License

This library is licensed under the MIT-0 License. See the LICENSE file.