Motion Objects Segmentation and Shadow Suppressing without Background Learning

Figure 6

Some results for the PETS’09 sequence with crowded people moving in the high-lights environment. The first row is the current frame; the second row is the segmented masks.