Abstract: Image inpainting is the process of filling in missing or damaged parts. Exemplar based image inpainting algorithm fills in the missing parts using information from the known part of the image. Scanning the whole image for finding the most similar patch is a very time consuming task. Also the algorithm doesn’t consider the depth information which harms the quality of the result. In this paper, we use depth information from the image depth map and this approach improves efficiency and quality of inpainted image. The paper includes experiment results and comparison with exemplar based inpainting algorithm.

Keywords: stereo image, inpainting, exemplar based, depth map, image completion

ACM Classification Keywords: I.4.4 Image Processing and Computer Vision - Restoration

Link:

EXEMPLAR BASED INPAINTING USING DEPTH MAP INFORMATION

Vahan Gevorgyan Vladimir, Aram Gevorgyan Vladimir, Gevorg Karapetyan Arakel

http://www.foibg.com/ijita/vol23/ijita23-03-p05.pdf