Lip Tracking for 3D Face Registration
Robert Anderson, Björn Stenger, Roberto Cipolla · 2013
This paper presents a method for 3D lip registration when tracking a face model from video and depth map data. The inner and outer lip contours are tracked in 2D using features obtained from Boosted Edge Learn-ing (BEL). Given the lip contours and a reconstructed depth map, the 3D mouth region of a mesh is registered at each frame independently, accounting for the self oc-clusions of the lips that occur during speech. Experi-ments show that this leads to significantly higher reg-istration accuracy in the perceptually important mouth region. The proposed method is also shown to have lower error than an Active Appearance Model based registration method. 1