作者: Jurgen Rurainsky , Peter Eisert
DOI: 10.1109/ICIP.2007.4379249
关键词:
摘要: We present our system for the capturing and analysis of 3D facial motion. A high speed camera is used as capture unit in combination with two surface mirrors. The mirrors provide additional virtual views face without need multiple cameras to avoid synchronization problems. use this motion a person's while speaking. Investigations these motions are presented rigid non-rigid analyzed. In order extract only deformation independent from head pose, we new simple approach separating named weight-compensated estimation (WCME). This weights data points according their influence desired model. also first results model-based analysis. Such can be animations achieve higher degree quality.