TianxingWu/OpenVHead
A vision-based head motion capture system that drives a 3D virtual character using facial landmark detection.

Velocity · 7d
+0.2
★ / day
Trend
→steady
star history
OpenVHead is a real-time virtual head control system for VTubers that uses computer vision to track head movement and facial expressions from webcam input. The system employs dlib for facial landmark detection and OpenCV for image processing, feeding the tracked data into a Unity 3D environment to animate a virtual character. Filters are applied to smooth motion and enhance robustness of the captured expressions.