AWS C++ rekognition example

This example demonstrates how to use AWS C++ SDK rekognition with focus on the face recognition.

Main features:

• List collections with related enrolled faces
• Create collections
• Enroll face to a collection given the face image file
• Recognize a face given the collection id and the face image file

Build

Pre requsite

1. The tool assumes that you have built and installed the AWS C++ SDK. Follow the link for instruction how to do it, https://docs.aws.amazon.com/sdk-for-cpp/v1/developer-guide/setup.html

2. We also assume that the AWS cli is installed and configured, for installation see https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-install.html. Configure AWS cli configure like:

aws configure

Build the tool

If you have been able to build and install the sdk, then simply run make:

make

Usage

Note!

All configurations and credentials are taken from you local AWS home. It's located in ~/.aws/.

List all enrolled faces for all Collections

To view all Collections enter following command:

Example

./aws-face-rekognition list

Create Collection

To create a Collection in AWS:

Example

./aws-face-rekognition create test-123

Enroll face

The following command will enroll a given image into AWS rekognition service:

Example

./aws-face-rekognition enroll 80 face123.jpeg

Recognize face

In order to recognize a face, pass the collection id and the face image like:

Example

./aws-face-rekognition recognize 80 face456.jpeg