IEEE Transactions on Circuits and Systems for Video Technology
Computer Vision and Image Understanding
Hi-index | 0.00 |
This paper presents a model-based technique for monocular tracking of the head pose using a non-calibrated camera. We use texture-mapped face images through the 3D head model as the data representation. The mapped data are compared to the model data via a similarity metric that expresses the likeness between the rendered and the reference images. The tracking is realized using a particle filter. In observation model we utilize rectangle features as the primary cue. The potential of our approach is demonstrated by tracking of the head pose on real videos.