Efficient HEVC decoder for heterogeneous CPU with GPU systems

Abstract

The High Efficiency Video Coding (HEVC) standard provides higher compression efficiency than other video coding standards but at the cost of increased computational load, which makes it hard to achieve real-time encoding/decoding of high-resolution, high-quality video sequences. In this paper, we investigate how Graphics Processing Units (GPUs) can be employed to accelerate HEVC decoding. GPUs are known to provide massive processing capability for throughput computing kernels, but the HEVC entropy decoding kernel cannot be executed efficiently on GPUs. We therefore propose a complete HEVC decoding solution for heterogeneous CPU+GPU systems, in which the entropy decoder is executed on the CPU and the remaining kernels on the GPU. Furthermore, the decoder is pipelined such that the CPU and the GPU can decode different frames in parallel. The proposed CPU+GPU decoder achieves an average frame rate of 150 frames per second for Ultra HD 4K video sequences when four CPU cores are used with an NVIDIA GeForce Titan X GPU.

Recent Advances on HEVC Inter-Frame Coding: From Optimization to Implementation and Beyond

“…Xiao introduced a fast HEVC encoding framework with multi-core CPUs and GPUs, in which GPUs are used to estimate MVs for each CU/PU partitions while CPUs are used to leverage the MC costs to speed up CU/PU partitions [103]. On the other side, hybrid GPU plus CPU architectures for HEVC decoder were also studied [104,105]. The performance of GPU-based implementations are listed in TABLE V. As can be seen, with the implementations on CPU/GPU, together with fast algorithms, HEVC encoding can be much accelerated.…”

Performance-efficient integration and programming approach of DCT accelerator for HEVC in MANGO platform

“…Previous research in this field rarely cover integration of the accelerators in heterogeneous systems containing different types of processing cores, from general processing nodes to heterogeneous nodes. In [4][5][6] CPU + GPU heterogeneous platform is used to accelerate HEVC decoder, either by deploying a subset of functions (IDCT and de-quantization) or by implementing entire decoder on GPU. In comparison with the HEVC encoder, the decoder is computationally much less demanding than HEVC encoder, which is why our focus is set on improving the encoding process.…”