Depth Map Re nement Using Reliability Based Joint Trilateral Filter

In this paper, we propose a re nement lter for depth maps. The lter convolutes an image and a depth map with a cross computed kernel. We call the lter joint trilateral lter. Main advantages of the proposed method are that the lter ts outlines of objects in the depth map to silhouettes in the image, and the lter reduces Gaussian noise in other areas. The e ects reduce rendering artifacts when a free viewpoint image is generated by point cloud rendering and depth image based rendering techniques. Additionally, their computational cost is independent of depth ranges. Thus we can obtain accurate depth maps with the lower cost than the conventional approaches, which require Markov random eld based optimization methods. Experimental results show that the accuracy of the depth map in edge areas goes up and its running time decreases. In addition, the lter improves the accuracy of edges in the depth map from Kinect sensor. As results, the quality of the rendering image is improved.