Opti-GVF Snake Model for Face Segmentation from Video Sequences.
Provat Biswas, Madhukar C. Pandit · Indian Conference on Computer Vision, Graphics and Image Processing · 2002
Face segmentation is one of the most important tasks in Model Based Video coding for Video Telephony and Video Conferencing applications. Since the face movement is usually different from the back ground movement, gradient of the optical flow field gives useful hints for the region of the image frame containing human face. In this work we have used a modified GVF-Snake (Gradient Vector Flow-Snake), where Gradient Vector Flow of the optical flow field is used as a component of the energy function to be minimized, to segment out the face from video sequence. GVF-Snake has a particular property that it can capture concave boundaries. Experiments with a number of test sequences give encouraging results.