Facial expression analysis robust to 3D head pose motion

A.C. Andres del Valle1, J.-L. Dugelay1
1Institut Eurécom, Sophia-Antipolis, France

Tóm tắt

Most face expression algorithms assume a front or 'near-to-front' head position. This assumption becomes an important limitation when studying input from real systems. We present a new approach to robustly determine face expressions independently of the head pose. Our analysis-synthesis cooperation, possible thanks to the use of a highly realistic 3D head model and the application of Kalman filtering to predict the user pose, permits us to correctly track the interesting face features. Adapting 'near-to-front' analysis techniques based on the predicted pose enables us to use such algorithms with moving speakers.

Từ khóa

#Robustness #Motion analysis #Algorithm design and analysis #Head #Performance analysis #Speech analysis #Image analysis #Predictive models #Image motion analysis #Teleconferencing

Tài liệu tham khảo

cosatto, 2000, Audio-visual selection for the synthesis of photo-realistic talking-heads, ICME 2000 10.1109/34.895976 andres del valle, 2001, Eye state tracking for face cloning, ICIP 2001 10.1109/ICME.2000.871438 chang, 2000, Vitual talk: A model-based virtual phone using a layered audio-visual integration, ICME 2000 10.1109/AFGR.2000.840678 10.1109/38.708562 10.1109/79.924885 dugelay, 2001, Analysis-synthesis cooperation for mpeg-4 realistic clone animation, Euroimage ICAV3D ogata, 2001, Model-based lip synchronization with automatically translated synthetic voice toward a multi-modal translation system, ICME 2001 10.1109/93.839309 andres del valle, 2002, Pose Coupling With Eye Movement Tracking