Memory-Aware Tiles Workload Balance through Machine-Learnt Complexity Reduction for HEVC

Abstract

This paper proposes a workload balancing algorithm aiming to speed up the HEVC parallel encoding using Tiles. Different from other literature works, the proposed solution uses static uniform tiling to avoid memory management difficulties that may emerge when dynamic tiling solutions are employed. The proposed algorithm relies on workload distribution history of previous frames to predict the workload distribution of the current frame. Then, it balances the workload among Tiles by employing a workload reduction scheme based on decision trees in the coding process. Experimental tests show that the proposed solution outperforms the standard uniform tiling and is competitive with related works in terms of speedup. Moreover, the solution optimizes resources usage in multiprocessing platforms, presents a negligible coding efficiency loss and reduces memory bandwidth usage by 9.34%.