Tracking Multiple Self-Occluding People using Dense Spatio-Temporal Motion Segmentation

Abstract. This master thesis discribes a new dense spatio-temporal motion segmentation algorithm with application to tracking of people in crowded environments. The algorithm is based on state-of-the-art motion and image segmentation algorithms. We specifically make use of a mean shift image segmentation algorithm and two graph based motion segmentation algorithms. The resulting motion segmentation is on the one hand accurate and on the other hand computationally efficient. In addition our method is capable of handling mutual occlusions. This shows that motion segmentation can efficiently be used to simultaneously detect, track and segment moving objects. We apply this to tracking people in surveillance videos, but the algorithm is not limited to this class of scenes.