6. Feature-Based Head Pose Tracking

We have studied an new approach to feature-based head tracking and pose estimation using a model-based analysis-by-synthesis scheme. The proposed algorithm employs a 3D head model, a feature-based pose estimation algorithm and texture mapping technique to dynamically produce templates for the feature tracking. In this way, the set of templates used for the matching is constantly updated with the pose changes, allowing the algorithm to track the features over a large range of head motion without loss of precision and error accumulation.