A NEW DEPTH-AIDED MULTIDIRECTIONAL DISOCCLUSION RESTORATION METHOD FOR DEPTH-IMAGE-BASED RENDERING

摘要

A new algorithm is proposed for restoring disocclusion regions in depth-image-based rendering (DIBR) warped images. Many disocclusion restoration methods have been proposed to solve this issue including layered depth image (LDI), pre-filtering methods, and post-processing methods, but they cause different types of annoying artifact for large holes with complex texture background. In this paper, a novel multidirectional extrapolation disocclusion restoration method is proposed to enhance visual quality for large hole with complex texture background. The proposed method features incorporation of a depth cue into texture estimation. Experiments on test stereoscopic sequence showed that our method has superiority in depth differentiation and geometrical loyalty in the restoration of warped images.