3D morphable model parameter estimation
AI'06 Proceedings of the 19th Australian joint conference on Artificial Intelligence: advances in Artificial Intelligence
Multimedia Tools and Applications
Hi-index | 0.00 |
This paper addresses the 3D tracking of pose and animationof the human face in monocular image sequencesusing deformable 3D models. For each frame, the proposedadaptation is split into two consecutive stages: global andlocal. In the fir st stage, the 3D pose of the face is recoveredusing a RANSAC-based technique involving both the consensusmeasure and the consistency with a statistical modelof a face texture. In the second stage, the local motion associatedwith some facial features is recovered using theconcept of the active appearance model search. Adaptationexamples demonstrate the feasibility and robustness of thedeveloped framework.