Overview

Learn how to use our endpoint to perform several common tasks in the biometric and document verification processes

The Progeny RESTful API https://be.progeny.tech/api/v3uses the HTTP POST and GET methods to create and update resources in our database.

The request body must be encoded as multipart/form-data.

All API methods return JSON response.

Making API calls and getting a response

To get a response from each of our endpoints you will first need to make a POST request to obtain a task Id. You then pass in the Id when forming your GET request to get a response.

We'll look at an example using the /liveness endpoint.

Get a task Id

POST https://be.progeny.tech/api/v3/services/face_liveness/

Request Body

Name	Type	Description
image*	String	The file path to an image of the user's face in PNG, JPEG, or JPG file format
extend_data	Boolean	Set to true to retrieve extended data in the liveness response

200: OK A task Id

{
    "id": "98008e56-cd6f-410c-9d8d-f9e620661395"
}

The next step is to pass in that Id into your GET request.

Perform a liveness check

GET https://be.progeny.tech/api/v3/services/task/?id={{task_id}}

Path Parameters

Name	Type	Description
id*	String	The Id you obtained from your POST request

200: OK The result of the liveness check

{
    "id": 2158,
    "status": "SUCCESS",
    "result": {
        "data": {
            "liveness": {
                "data": {
                    "error": "Failed to detect face",
                    "error_code": "FACE_NOT_FOUND"
                },
                "result": false
            },
            "issues": {
                "liveness": {
                    "FACE_NOT_FOUND": "Failed to detect face"
                }
            }
        },
        "status": 200
    }
}

A quick word about images

Please be aware that our algorithm requires the submitted picture to simulate real-life scenarios. When testing but also in providing guidance to your end-users, please bear in mind the following points:

• Images should be taken with a device under real-life lightning conditions and without filters

• Avoid using artificially generated selfies, filtered, or enhanced images

• Similarly, avoid using images where the subject is wearing sunglasses, has long or bushy facial hair that sticks out a lot in front of the face, or has a significantly distorted face

• Images taken under bad light conditions, with high motion blur, or excessive light will not provide the most accurate results

Furthermore, images included in each API request must conform to the following requirements:

• PNG, JPEG, or JPG file format and as uncompressed as possible. Do not go below 70%

• Between 480x480 and 4096x4096 pixels. For the best results, we recommend a maximum height of 1080px

• No larger than 2MB

• Face size should be no lower than 224 pixels

• The minimum resolution between the subject's eyes should be no lower than 85 pixels

• Orientation:

  • Face pitch (equivalent to looking up and down) should be between -25 to 25 degrees

  • Yaw angle (equivalent to looking left and right) should be between -25 to 25 degrees

  • Face roll angle (equivalent to tilting left and right) from -35 to 35 degrees

Endpoints - Quick View

Endpoint	Description
/accounts/token/obtain	authenticate and obtain an access token
/accounts/token/refresh	Can be used to obtain new token without the need to send username/password again See: https://www.oauth.com/oauth2-servers/making-authenticated-requests/refreshing-an-access-token/
/services/register_face/	Register an image of a person's face with selected provider to be used later for Authentication
/services/authenticate_face/	Given an image of a person's face and an ID (provided by register_face API), it determine whether the registered face match the provided image or not
/services/liveness/	Perform a standalone liveness check against an image
/services/ocr_card/	Detect and extract text from a card (ID, Driver License, ..)
/services/ocr_passport/	Same as ocr_card but for reading passports
/services/kyc_card/	A combination ocr_card, liveness APIs plus comparing ID/Drive License picture against the one provided in the request
/services/kyc_passport	Same as kyc_card but for passports