Target Contour Recovering for Tracking People in Complex Environments

Figure 9

Some examples of occluded images based on MPEG-7 dataset. (a) is selected as the template image. (b), (c), (d), and (e) are the occluded images by single object under different occlusion region. (f) is the occluded image by two objects. Images in (g) are occluded by three objects. The red line indicates the occlusion region.