Abstract
Video inpainting is a process of repairing the damaged areas of a video or removing any desired part of a video. Dealing with such problems requires a sturdy image inpainting algorithm along with a regeneration technique for filling the missing parts of a video sequence recorded from a static camera. Many automatic techniques for video inpainting are available but most of them are computationally intensive and fail to repair the damaged areas. Also, from the remaining video inpainting algorithms a masked video sequence has to be provided. To overcome this problem, inpainting process is carried out by using the background registration method which is proposed in this paper. The video is first converted into distinct image frames and the first frame is registered. Second, the edges of an object to be removed are detected by comparing the registered frame with each succeeding frame of the video. Next, a masked frame is generated for each time frame. Then the inpainting process is performed separately for each and every time frame of the images. Next, these processed image frames are displayed sequentially, so that it appears as a video.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
