Object Contour Tracking Using Foreground and Background Distribution Matching
Quen Outaouais · 2009
In this paper, we propose an effective approach for track- ing distribution of objects. The approach uses a competition between a tracked objet and background distributions using active contours. Only the segmentation of the object in the first frame is required for initial- ization. The object contour is tracked by assigning pixels in a way that maximizes the likelihood of the object versus the background. We imple- ment the approach using an EM-like algorithm which evolves the object contour exactly to its boundaries and adapts the distribution parameters of the object and the background to data.