Face recognition does not always need a cloud service. HelloFace SDK brings face processing to the device, giving developers native integration options for Android, iOS, HarmonyOS, Windows, and Linux.

What the SDK provides

  • Face detection and feature extraction.
  • Guided face capture.
  • Action-based and passive liveness detection.
  • 1:1 face verification.
  • 1:N and M:N face search with local database management.

The SDK processes face images and recognition features locally, without sending them to a remote face-recognition service.

HelloFace product overview and application scenarios

This illustration comes from the project's public README. The displayed match value is illustrative, not a recognition-accuracy benchmark or a hardware test result.

The Android demo is now on Google Play

The Android demo is now available on Google Play after completing Google Play's publication review. That review process includes checks against applicable app safety and privacy policies. Learn about Google Play's review process.

Download the Android demo on Google Play

The SDK's on-device face processing and the Android app's overall data practices should be considered separately. The Google Play Data safety section provides the app's current developer disclosures.

Linux and embedded devices

For Linux developers, HelloFace provides x86_64 and ARM64 binaries, a public C ABI, command-line examples, and a customizable desktop demo. The SDK can be integrated into applications without a desktop environment; the supplied graphical demo requires X11 or XWayland.

For embedded Linux projects, the demo provides display rotation, touch-friendly controls, and RGB/IR paired-search integration for supported hardware. Camera configuration, calibration, and performance should be validated on the target device.

Native mobile integration

HarmonyOS developers can integrate through ArkTS interfaces and a native HAR package. Android and iOS have their own native integration guides and demos. Public demo integration code and platform guides are available on GitHub; the SDK core is distributed as binaries.

The public 0.8.11 release is Beta and intended for testing. The current models do not support children under 12 and must not be used in financial scenarios. Face-search deployments require clear camera images; outdoor deployments require camera wide dynamic range above 100 dB. Windows/Linux desktop and camera workflows still require validation with real cameras; mobile apps require their own application signing.

Explore the examples, try the Android demo, and share your integration feedback: