Motion Objects Segmentation and Shadow Suppressing without Background Learning

Figure 3

Some results for the intelligent room sequence. The first row is the current frame; the second row is the segmented masks.