Abstract
The Multi-View extension of High Efficiency Video Coding (MV-HEVC) has improved the coding efficiency of multi-view videos, but this comes at the cost of the extra coding complexity of the MV-HEVC encoder. This coding complexity can be reduced by efficiently reducing time-consuming encoding operations. In this work, we propose two methods to reduce the encoder complexity. The first one is Early Coding unit Splitting (ECS), and the second is the Efficient Reference Picture Selection (ERPS) method. In the ECS method, the decision of Coding Unit (CU) splitting for dependent views is made on the CU splitting information obtained from the base view, while the ERPS method for dependent views is based on selecting reference pictures on the basis of the temporal location of the picture being encoded. Simulation results reveal that our proposed methods approximately reduce the encoding time by 58% when compared with HTM (16.2), the reference encoder for MV-HEVC.
Original language | English |
---|---|
Article number | 454 |
Journal | Symmetry |
Volume | 11 |
Issue number | 4 |
DOIs | |
Publication status | Published - 1 Apr 2019 |
Externally published | Yes |
Keywords
- Coding Unit
- MV-HEVC
- Prediction Unit
- Rate Distortion
- Reference picture
- Video coding