SentiMask SDK
3D face tracking for augmented reality apps and digital characters control
SentiMask is designed for development of augmented reality applications, which use real-time 3D face tracking technologies for motion capture and controlling 3D digital character's facial expressions or mapping animated avatars on user's face. The technology works with regular cameras and common PC or smartphones.
Available as a software development kit that provides for the development of 3D face tracking systems for Microsoft Windows, Android, iOS, macOS and Linux.
Specifications and Usage Recommendations
- Only one face in a frame is processed. If there are more than one face in a frame, the largest one is processed.
- 32 pixels is the recommended minimal distance between eyes for a face on image or video stream to perform face detection reliably. 64 pixels or more recommended for better face detection results. Note that this distance should be native, not achieved by resizing an image.
-
The SentiMask engine has certain tolerance to face posture:
- head roll (tilt) – ±180 degrees;
- head pitch (nod) – ±25 degrees from frontal position;
- head yaw (bobble) – ±35 degrees from frontal position.
- Face detection and facial features estimation in a video frame is performed in 8 milliseconds on a PC with Intel Core i7-4790 processor running at 3.6 GHz clock rate.
