Robust Exemplar based Object Removal in Video

Sreelekshmi Das, Reeba .R

Abstract: Image completion or image inpainting is the technique that automatically completes or restores removed areas in an image. Exemplar-based image inpainting technique can be used to remove objects from an image and fill the hole area with matching background content. This paper presents an exemplar-based video inpainting mechanism that restores the area of the removal object, and the mechanism can be further employed to extract the back ground of videos. Firstly, video is converted in to frames and object tracking in each frame using normalized cross correlation. Exemplar-based inpainting methods iteratively search the source region and fill the missing or damaged region, i.e., target region, with the most similar patch in the source region. With segmentation map, the proposed methods automatically select the parameter values for the robust priority function and reduce the search region.

Keywords: 2D Normalized cross correlation; Exemplar-based inpainting; Segmentation map; Robust priority function