Index Faces (Rekognition)

API reference

curl -X POST "https://__mock__/index_faces" \  -H "Content-Type: application/json" \  -d '{    "image": "string",    "bucket": "string",    "image_key": "string",    "external_image_id": "string"  }'

{
  "success": true,
  "message": "string",
  "data": {}
}

{
  "msg": "string"
}

{
  "detail": {},
  "message": "string"
}

Overview

Call POST /index_faces with Authorization: Bearer <token> and application/json. Required body fields are image (base64), bucket, image_key, and external_image_id. The server must have AWS Rekognition and related configuration enabled.

Key features

• Searchable gallery: Indexes a face for later search and duplicate checks when your deployment uses the Rekognition collection.

Implementation

Step 1: Build the JSON payload

Include required fields image, bucket, image_key, and external_image_id as described in the OpenAPI schema.

Step 2: Send the request from your backend

POST /index_faces HTTP/1.1
Host: <machine-learning-api-base-url>
Authorization: Bearer <token>
Content-Type: application/json

Step 3: Handle the response

On 200 OK, read success, message, and data from the JSON body. On 400, read msg.

Error handling

Benefits

• Centralizes face indexing for downstream face-search workflows.

Next steps

• Delete faces from collection
• Facematch
• API Suite index