GPU를 이용한 깊이영상 기반 고속 스테레오 비디오 생성 방법

In this paper, we propose a fast rendering method based on the depth image to generate a virtual view image in real-time using a graphic processor unit (GPU) for a 3D broadcasting system. In order to take advantage of the single instruction multiple data structure of GPU, we collect independent data sets and use high speed memories. As a result, we can generate virtual view images 15 times faster than the normal CPU processing.