Title :
A DIBR method based on inverse mapping and depth-aided image inpainting
Author :
Cong Cheng ; Ju Liu ; Hui Yuan ; XiaoHui Yang ; Wei Liu
Author_Institution :
Sch. of Inf. Sci. & Eng., Shandong Univ., Jinan, China
Abstract :
Depth-image-based rendering (DIBR) has attracted more and more attention as a method of virtual view rendering in three-dimensional video (3DV) applications. Many efforts have been dedicated to realize high-quality rendering, ranging from pre-process of the reference depth images to post-process of the virtual texture images. A DIBR method based on inverse mapping and depth-aided image inpainting technique is proposed in this paper. In the proposed method, inverse mapping rendering technique is adopted to realize the 3D warping instead of traditional forward mapping; and a modified image inpainting technique in which depth information is considered is employed to inpaint the disocclusion regions. Besides, morphological processing is performed on the reference depth image to avoid the artifacts along the disocclusion areas. Experimental results demonstrate that the proposed method can obtain higher peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) in comparison to the previous methods, while producing satisfactory visual effect.
Keywords :
image texture; rendering (computer graphics); video signal processing; 3D warping; 3DV applications; DIBR method; PSNR; SSIM; depth-aided image inpainting; depth-image-based rendering; disocclusion regions; high-quality rendering; inverse mapping rendering technique; modified image inpainting technique; morphological processing; peak signal-to-noise ratio; reference depth image preprocessing; structural similarity index; three-dimensional video application; virtual texture image post-processing; virtual view rendering; Cameras; Educational institutions; Multimedia communication; PSNR; Rendering (computer graphics); TV; Three-dimensional displays; 3DV; depth image based rendering; image inpainting; inverse mapping; view synthesis;
Conference_Titel :
Signal and Information Processing (ChinaSIP), 2013 IEEE China Summit & International Conference on
Conference_Location :
Beijing
DOI :
10.1109/ChinaSIP.2013.6625394