This video demonstrates the MCMCDA based head tracking system, which runs at 25fps on 1920x1080 video using a standard desktop computer. The system is capable of obtaining stable head images and is robust to temporary occlusions.
Examples showing results of the unsupervised coarse gaze estimation. The gaze direction estimates are the result of fully unsupervised learning - the system was not provided with any example images to link gaze directions with image appearances. The gaze directions were learned by simultaneously optimising both dynamic and appearance models using a Conditional Random Field (CRF) in combination with a forest of randomised decision trees.