3D Video Compression Based on High Efficiency Video Coding

With the advent of autostereoscopic displays, questions rise on how to efficiently compress the video information needed by such displays. Additionally, for gradual market acceptance of this new technology it is valuable to have a solution offering forward compatibility with stereo 3D video as it is used nowadays. In this paper, a multiview compression scheme making use of the efficient single-view coding tools used in High Efficiency Video Coding (HEVC) is provided. Although efficient single view compression can be obtained with HEVC, a multiview adaptation of this standard under development is proposed, offering additional coding gains. On average, for the texture information, the total bitrate can be reduced by 37.2% compared to simulcast HEVC. For depth map compression, gains largely depend on the quality of the captured content. Additionally, a forward compatible solution is proposed offering the possibility for a gradual upgrade from H.264/AVC based stereoscopic 3D systems to an HEVC-based autostereoscopic environment. With the proposed system, significant rate savings compared to Multiview Video Coding (MVC) are presented(1).