Adaptive fractional motion and disparity estimation skipping in MV-HEVC |
| |
| Affiliation: | 1. School of Intelligent Mechatronics Engineering, Sejong University, South Korea;2. School of Computer Science and Engineering, Kyungpook National University, South Korea;1. Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China;2. Zhejiang Wanli University, Ningbo, China;3. University of Thessaly, Volos, Greece;4. National Computer Network Emergency Response Technical Team/Coordination Center of China, China |
| |
| Abstract: | MV-HEVC can efficiently compress multiview video data captured from different viewpoints. To achieve high coding efficiency, it consists of not only inter coding but also interview coding. The inter coding includes a motion estimation (ME) process that reduces temporal redundancies between consecutive frames, and the interview coding performs a disparity estimation (DE) that reduces interview redundancies between neighboring views. As a result, MV-HEVC needs high encoding complexity to perform both ME and DE. In order to reduce the complexity, this paper proposes an adaptive fractional ME and DE skipping method in a partitioned inter prediction unit (PU) mode, based on a result of a 2 N × 2 N inter PU coding. Experimental results show that the proposed method efficiently reduces the encoding complexity with negligible coding loss, compared to conventional methods. |
| |
| Keywords: | Encoding complexity Disparity estimation (DE) Motion estimation (ME) Interview coding MV-HEVC |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
