Depth-aided Exemplar-based Hole Filling for DIBR View Synthesis
|Name||Depth-aided Exemplar-based Hole Filling for DIBR View Synthesis|
Quality of synthesized view by Depth-Image-Based Rendering (DIBR) highly depends on hole filling, especially for synthesized view where large disocclusion exists. Thus, many hole filling methods are proposed to improve the synthesized view quality. Inpainting is the most popular approach to recover the disocclusion regions. However, the conventional inpainting either makes the disoccluded regions blurred via diffusion or propagates the foreground information to the disoclusion regions. Annoying artifacts are created in the view synthesis process. This paper proposes a depth-aided exemplar- based inpainting method for large disoclusion recovery. It consists of twotwo processes, warped depth map hole filling and warped color image hole filling. Since depth map can be considered as a grey-scale image without texture, it is much easier to be filled. Disoccluded regions of color image are predicted based on its associated depth map information. Regions with texture lying around the background have higher priority to be filled than other regions and disoccluded regions are filled by propagating the background textures through the exemplar-based inpainting. Thus artifacts created by diffusion or using foreground information for prediction can be eliminated. Experimental results show texture can be recovered in large hole regions and the proposed method has better visual quality compared to existing methods.
|ieee paper year||2013|