Moving Target Removal in Video Sequence Using Boundary Tracking

Moving target removal as one of the video special effect techniques refers to removing a foreground moving object from background and filling in the missing regions in a visually consistent manner. Usually for fill-in, we have to manually create masks for moving targets frame by frame by human intervention. This paper introduces a convenient algorithms to implement moving target removal from constrained complex background in a video sequence. We start from a given mask of the moving target in the first frame, and propose a new method which combines frame difference with boundary tracking to automatically generate moving target masks for other frames, then use image synthesis and PDE-based inpainting to repair the region of the moving target in the first frame. Finally, we fill the mask regions of moving target in other frames to finish the removal of moving target in video sequence.