作者: Jan Hanca , Adrian Munteanu , Peter Schelkens
DOI: 10.1117/12.2002106
关键词:
摘要: Efficient depth-map coding is of paramount importance in next generation 3D video applications such as 3DTV and free viewpoint video. In this paper we propose a novel intra depth map system employing optimized segmentation procedures suitable for maps, followed by lossy or lossless contour techniques. mode, our method performs Hausdorff-distance constrained coding, which the distance between actual decoded contours upper-bounded user-defined bound. The trade-off location accuracy performance analyzed. Experimental results show that, on average, outperforms should be considered all segmentation-based systems. comparison against JPEG-2000 shows that proposed viable alternative light maps.