FACE_GUARD_PRO integration process,DEVICE_GUARD_PRO integration process - ID Verification

FACE_GUARD_PRO is a solution that detects potential threats in the authentication environment. It is designed for customers who cannot fully integrate the ID Verification product. FACE_GUARD_PRO detects device threats through in-depth analysis from the Tongyi large language model.

Note The Face Guard product solution has been renamed to Device Guard, but the product code remains unchanged. For more information, see the announcement [Announcement] Renaming of the Face Guard product solution.

Integration overview

Important

Applications that integrate FACE_GUARD_PRO require a mature facial recognition algorithm and a threat policy operations system. Contact your business manager to assess whether your business scenario is a good match.
FACE_GUARD_PRO detects device threats on various clients. However, each client platform has unique system environment and permission requirements. For more information, see the Usage Notes section in the integration document for the corresponding client.

The FACE_GUARD_PRO workflow collects information from a client device and sends it to the server for detection. You can use the server-side query API to retrieve detailed threat labels.

This solution requires an application server (server) and a client. The following time series chart shows the entire FACE_GUARD_PRO workflow.

As shown in the preceding figure, the complete FACE_GUARD_PRO flow has two stages:

The first stage, collection and reporting, occurs only on the client.
Note
The connection to the FACE_GUARD_PRO server is automatically handled by the software development kit (SDK). Therefore, the client must be able to connect to Alibaba Cloud servers.
In the second stage, the application server retrieves the detection result, processes it, and returns the final outcome to the client.

If you use a facial recognition service with FACE_GUARD_PRO, note the following two call timings on the client:

Timing 1: The user is about to start facial authentication, but before the camera is activated. You can call the FACE_GUARD_PRO SDK initialization API when the user enters the guide page, or earlier. The sooner the call is made, the better.
Timing 2: After the user completes all facial authentication actions, including activating the camera and scanning their face. Ensure there is a delay of at least 3 seconds between this timing and Timing 1.
When calling getDeviceToken, pass in the bizId, which is the ID for the current facial scan authentication. This action binds the current deviceToken to the unique business ID for the facial scan. When the server queries the threat result, it can pass in the bizId to verify the association.