Yongfei Zhang, Rui Fan, Chao Zhang, G. Wang, Zhe Li
{"title":"DSP上HEVC编码的SIMD加速","authors":"Yongfei Zhang, Rui Fan, Chao Zhang, G. Wang, Zhe Li","doi":"10.1109/APSIPA.2017.8282310","DOIUrl":null,"url":null,"abstract":"As the new generation video coding standard, High Efficient Video Coding (HEVC) significantly improves the video compression efficiency, which is however at the cost of a far more computational payload than the capacity of real-time video applications and general purpose processors. In this paper, we focus on the SIMD-based fast implementation of the HEVC encoder over modern TI Digital Signal Processors (DSPs). We first test the DSP-based HEVC encoder and indentify the most time-consuming encoding modules. Then SIMD instructions are exploited to improve the parallel computing capacity of these modules and thus speed up the encoder. The experimental results show that the proposed implementations can significantly improve the encoding speed of the DSP-based HEVC encoder, with a speedup ratio of 8.38-87.32 over the original C-based encoder and 1.59–6.56 over o3-optimization enabled encoder.","PeriodicalId":142091,"journal":{"name":"2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":"{\"title\":\"SIMD acceleration for HEVC encoding on DSP\",\"authors\":\"Yongfei Zhang, Rui Fan, Chao Zhang, G. Wang, Zhe Li\",\"doi\":\"10.1109/APSIPA.2017.8282310\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"As the new generation video coding standard, High Efficient Video Coding (HEVC) significantly improves the video compression efficiency, which is however at the cost of a far more computational payload than the capacity of real-time video applications and general purpose processors. In this paper, we focus on the SIMD-based fast implementation of the HEVC encoder over modern TI Digital Signal Processors (DSPs). We first test the DSP-based HEVC encoder and indentify the most time-consuming encoding modules. Then SIMD instructions are exploited to improve the parallel computing capacity of these modules and thus speed up the encoder. The experimental results show that the proposed implementations can significantly improve the encoding speed of the DSP-based HEVC encoder, with a speedup ratio of 8.38-87.32 over the original C-based encoder and 1.59–6.56 over o3-optimization enabled encoder.\",\"PeriodicalId\":142091,\"journal\":{\"name\":\"2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)\",\"volume\":\"71 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"7\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/APSIPA.2017.8282310\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/APSIPA.2017.8282310","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
摘要
高效视频编码(High efficiency video coding, HEVC)作为新一代视频编码标准,显著提高了视频压缩效率,但其代价是计算量远远超过实时视频应用和通用处理器的能力。在本文中,我们重点研究了基于simd的HEVC编码器在现代TI数字信号处理器(dsp)上的快速实现。我们首先测试了基于dsp的HEVC编码器,并确定了最耗时的编码模块。然后利用SIMD指令来提高这些模块的并行计算能力,从而提高编码器的速度。实验结果表明,所提出的实现方案可以显著提高基于dsp的HEVC编码器的编码速度,比原来基于c的编码器的加速比为8.38 ~ 87.32,比支持o3优化的编码器的加速比为1.59 ~ 6.56。
As the new generation video coding standard, High Efficient Video Coding (HEVC) significantly improves the video compression efficiency, which is however at the cost of a far more computational payload than the capacity of real-time video applications and general purpose processors. In this paper, we focus on the SIMD-based fast implementation of the HEVC encoder over modern TI Digital Signal Processors (DSPs). We first test the DSP-based HEVC encoder and indentify the most time-consuming encoding modules. Then SIMD instructions are exploited to improve the parallel computing capacity of these modules and thus speed up the encoder. The experimental results show that the proposed implementations can significantly improve the encoding speed of the DSP-based HEVC encoder, with a speedup ratio of 8.38-87.32 over the original C-based encoder and 1.59–6.56 over o3-optimization enabled encoder.