Depth Map Boundary Enhancement Using Random Walk

Depth information is essential for depth image-based rendering (DIBR), which is one of the rendering processes for the virtual view using a color image and its corresponding depth map. Although there are several depth estimation methods, more accurate depth estimation is still required. Since inaccurate depth information along object boundaries causes serious rendering errors, depth boundary information is very important. In this paper, we propose a new depth map filtering algorithm to solve the boundary mismatch problem. In order to preserve depth boundaries, we smooth depth values of the non-boundary region using the random walk probability. After calculating a similarity measure between the current pixel and its neighboring pixels using the random walk probability model, we have applied a weighted averaging filter based on the random walk probability. Finally, we have obtained a depth map with higher quality compared to other methods. Experimental results showed enhancement of depth boundaries.