دانلود رایگان مقاله نسخه رمزگذار موازی همزمان و ناهمزمان HEVC با GOP

Abstract

In this paper, we focus on applying parallel processing techniques to HEVC encoder in order to significantly reduce the computational power requirements without disturbing its coding efficiency. So, we propose several, synchronous and asynchronous, parallelization approaches working at a coarse grain parallelization level, based on the Group Of Pictures (GOP), which we call GOP-based level. GOP-based approaches encode simultaneously several groups of consecutive frames. Depending on how these GOPs are conformed and distributed it is critical to obtain good parallel performance . The results show that near ideal efficiencies are obtained using up to 10 cores. Furthermore, when the computational load is unbalanced, the asynchronous versions outperform the synchronous ones. The parallel algorithms developed in this work support all standard coding modes proposed by the reference software.

6. Conclusions

In this paper we have proposed several parallel algorithms of the HEVC video encoder. These algorithms are based on a coarser grain parallelization approach with the organization of video frames in Group Of Pictures (GOP) and different GOP allocation schemes. A good parallel behavior has been shown in the experiments reported, which were obtained using a multicore platform. However the developed algorithms are able to run on distributed memory architectures since a coarser grain parallelization has been used. We have presented results using the different encoding modes proposed by the reference software, analyzing its performance. After implementing the algorithms in the HEVC software some experiments were performed showing interesting results as (a) GOP organization determines the final coding performance, being the best approach Option S-IV (AI mode) algorithm when comparing both sequential and parallel versions in terms of speed-up/efficiency; (b) although Option S-III algorithm introduces a bit rate overhead as the number of processes increases, the overall parallel performance and the improvements in PSNR make it a good approach when LB, LP or RA coding modes are demanded; (c) Option S-V algorithm offers similar features than Option S-III but with the ability to update the bit stream during encoding process with data obtained from all processes, not just from the root process; and (d) asynchronous versions of S-I, S-IV y S-V algorithms were provided showing slightly lower encoding times with the ability of load-balancing the input workload among the available processes.