Free viewpoint video synthesis using multi-view depth and color cameras

Abstract

In this paper, we propose an approach for generating free viewpoint videos based on multiple depth and color cameras to resolve issues encountered with traditional color cameras techniques. Our system is based on consumer products such as Kinect that does not provide satisfying quality in terms of resolution and noise. Our contribution is then to propose a full pipeline for enhancing the depth maps and finally improving the quality of the novel viewpoint generated.