Face Recognition System

Free version of plugin

Important:

After you build your project for Windows you must copy the library libtensorflowlite.so from the folder Assets/FaceRecognitionSystem/Plugins/x86_64 and paste it to the folder Plugins/x86_64 of the built project (near the library FRS.dll).

• Face recognition solution, including face detection (using 6 keypoints: left eye, right eye, nose, mouth, left ear, right ear), face alignment, face normalization, face feature extraction and face search;

• It recognizes faces quite accurately (up to 99.5% on LFW dataset);

• Age recognition;

• Gender recognition;

• Emotion recognition;

• Head pose estimation;

• Face mesh keypoints detection;

• It works offline, in real time;

• It includes a demo scene with an example of face search and recognition on a photo, a video, a video from a web-camera;

• Simple interface, low learning cost, can be applied to actual projects immediately, especially suitable for face attendance or face login authentication.

Supported platforms:

• Windows 10, Windows 7 Service Pack 1 (x64);

• Android (armeabi-v7a or arm64-v8a).

Face Recognition System uses OpenCV 4.5.0 under Apache 2.0 License; see Third-Party Notices.txt file in package for details.