Selfie Face Comparison

Overview

Compare face with face in ID card and decide whether they are from the same person

Image Requirements

Format : JPG (JPEG), PNG Size : between 112 x 112 and 40964096 (pixels) Minimal size of face : The bounding box of a detected face is a square. The minimal side length of a square should be no less than 112 pixels.

Return Values

Field	Type	Description
request_id	String	Unique id of each request
confidence	Float	ndicates the similarity of two faces, a floating-point number with 2 decimal places between [0,100]. Higher confidence indicates higher possibility that two faces belong to same person. Note: if no face is detected within the image uploaded, this string will not be returned.
thresholds	Object	A set of thresholds including 3 floating-point numbers with 2 decimal places between [0,100]. If the confidence does not meet the "err_01" threshold, it is highly suggested that the two faces are not from the same person. While if the confidence is beyond the "err_0001" threshold, there's high possibility that they are from the same person. err_01: confidence threshold at the 0.1% error rate; err_001: confidence threshold at the 0.01% error rate; err_0001: confidence threshold at the 0.001% error rate; Note: seeing that thresholds are not static, there's no need to store values of thresholds in a persistent form, especially not to compare the confidence with a previously returned threshold. If no face is detected within the image uploaded, this string will not be returned.
time_used	Float	Time that the whole request takes. Unit: second
error_message	String	This string will not be returned unless request fails. For more details, please see the following section on error message.

Response Definition

HTTP status	Error Message	Description
200	CAN_NOT_DETECT_FACE	The image which don't have face.
200	GOT_ONE_FACE	The image which has only one face.
400	IMAGE_ERROR_UNSUPPORTED_FORMAT	The image which indicates can not be resolved. The file format may not be supported or the file is damaged.

Compare face with face in ID card and decide whether they are from the same person

POST https://apis.aigen.online/aiface/selfie-doc-compare/v1

Request Body

Name	Type	Description
image*	String	​Base64 encoded binary data of image

200: OK

{
  "request_id": "string",
  "confidence": 100,
  "thresholds": {
    "err_01": 45.92,
    "err_001": 54.36,
    "err_0001": 59.71
  },
  "time_used": 0,
  "error_message": "error_message"
}

422: Unprocessable Entity

{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Python

import requests
import json

api = "https://apis.aigen.online/aiface/selfie-doc-compare/v1"
headers = {"x-aigen-key": "<key>", "content-type": "application/json"}
data = json.dumps({
    "image": "<base64_string>"
})

res = requests.post(api, data=data, headers=headers)
print(res.json())
​

Nodejs

const axios = require("axios");

const api = "https://apis.aigen.online/aiface/selfie-doc-compare/v1";
const headers = {
  "x-aigen-key": "<key>",
};
const data = { 
  image: "<bas64_string>"
};

axios
  .post(api, data, { headers: headers })
  .then((res) => {
    console.log(res.data);
  })
  .catch((err) => {
    console.error(err.response.data);
  });