
Rate control (RC) plays a critical role in video coding for regulating the coded bit‐stream to achieve the target bitrate, which involves two basic steps: bit allocation (BA) and quantisation parameter selection. Here, the former is a key element to directly affect the performance of RC. So, in this study, an efficient frame‐level BA algorithm is proposed for the latest video coding standard H.265/high‐efficiency video coding (HEVC). First, by analysing the optimisation problem of the frame‐level BA, two factors have been found to be related to the frame‐level BA, which are inter‐frame dependency and the target bitrate. Then, a measure of inter‐frame dependency is put forward to reflect the correlation among consecutive frames. Finally, according to the results of allocation weights through the dynamic programming method, the relationships among the allocation weights and the two factors are investigated, and based on which the efficient BA model is established. Experimental results show that in comparison with the original RC scheme in HM‐14.0, the proposed frame‐level BA algorithm reduces, on average, BD‐rate by 5.19% with negligible encoding time increase. In addition, the proposed algorithm also achieves higher GOP‐level RC accuracy compared with the original RC scheme in HM.
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 14 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Top 10% | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Top 10% | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Top 10% |
