AUGMENTED REALITY
Wearables
Perspective registration
Video Augmentation
The WristCam

RECENT PUBLICATION
Patents
Journals
Conferences

RESOURCES
Media Tech Resources

RETURN TO:
Userport homepage
York homepage

Our object-based video coding work inspired the use of deformable 2D and 3D head models for enhancement of video images. Given a degraded sequence of pictures (taken, for example, from a security camera), we fit a structural model to a person's face in the scene, then bring the different faces into correspondence and use temporal filtering to remove noise while sharpening in the spatial domain. This cleans up the appearance of faces, making them, potentially, more recognizable from video sequences.

Images generated by Xiaoming Ping