HEVC编码器中运动估计的高吞吐量硬件实现

A. Medhat, A. Shalaby, M. Sayed
{"title":"HEVC编码器中运动估计的高吞吐量硬件实现","authors":"A. Medhat, A. Shalaby, M. Sayed","doi":"10.1109/MWSCAS.2015.7282040","DOIUrl":null,"url":null,"abstract":"This paper presents a highly parallel motion estimation architecture for High Efficiency Video Coding (HEVC) encoder. The proposed architecture has 16 processing units operating in parallel to calculate the sum of absolute difference values of all possible variable prediction block sizes. Hence, it calculates the bit cost regarding every partition in order to find the best matching candidate in terms of bit cost. The proposed unit processes block sizes from 4×4 up to 64×64. The proposed architecture was prototyped, simulated and synthesized using 65nm TSMC CMOS technology. At 720 MHz clock frequency, the proposed architecture processes 2K (1920×1080) resolution at 30 fps with ±27 (55×55) pixel search range using full search algorithm. Moreover, the proposed architecture is a flexible one and it can be used with different search algorithms to process higher resolutions such as 4K (3840×2160) resolution with 30 fps rate. To the best of our knowledge, the proposed architecture is one of the first ASIC motion estimation architectures in the literature for HEVC.","PeriodicalId":216613,"journal":{"name":"2015 IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"17","resultStr":"{\"title\":\"High-throughput hardware implementation for motion estimation in HEVC encoder\",\"authors\":\"A. Medhat, A. Shalaby, M. Sayed\",\"doi\":\"10.1109/MWSCAS.2015.7282040\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents a highly parallel motion estimation architecture for High Efficiency Video Coding (HEVC) encoder. The proposed architecture has 16 processing units operating in parallel to calculate the sum of absolute difference values of all possible variable prediction block sizes. Hence, it calculates the bit cost regarding every partition in order to find the best matching candidate in terms of bit cost. The proposed unit processes block sizes from 4×4 up to 64×64. The proposed architecture was prototyped, simulated and synthesized using 65nm TSMC CMOS technology. At 720 MHz clock frequency, the proposed architecture processes 2K (1920×1080) resolution at 30 fps with ±27 (55×55) pixel search range using full search algorithm. Moreover, the proposed architecture is a flexible one and it can be used with different search algorithms to process higher resolutions such as 4K (3840×2160) resolution with 30 fps rate. To the best of our knowledge, the proposed architecture is one of the first ASIC motion estimation architectures in the literature for HEVC.\",\"PeriodicalId\":216613,\"journal\":{\"name\":\"2015 IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"17\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/MWSCAS.2015.7282040\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MWSCAS.2015.7282040","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 17

摘要

提出了一种用于高效视频编码(HEVC)编码器的高度并行运动估计结构。所提出的架构有16个处理单元并行运行,以计算所有可能的可变预测块大小的绝对差值之和。因此,它计算每个分区的比特成本,以便根据比特成本找到最佳匹配候选。建议的单元处理从4×4到64×64的块大小。采用台积电65nm CMOS技术对所提出的架构进行了原型、仿真和合成。在720 MHz时钟频率下,该架构使用全搜索算法以30 fps的速度处理2K (1920×1080)分辨率,搜索范围为±27 (55×55)像素。此外,所提出的架构是一种灵活的架构,可以与不同的搜索算法一起使用,以处理更高的分辨率,例如4K (3840×2160)分辨率和30fps速率。据我们所知,所提出的架构是HEVC文献中第一个ASIC运动估计架构之一。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
High-throughput hardware implementation for motion estimation in HEVC encoder
This paper presents a highly parallel motion estimation architecture for High Efficiency Video Coding (HEVC) encoder. The proposed architecture has 16 processing units operating in parallel to calculate the sum of absolute difference values of all possible variable prediction block sizes. Hence, it calculates the bit cost regarding every partition in order to find the best matching candidate in terms of bit cost. The proposed unit processes block sizes from 4×4 up to 64×64. The proposed architecture was prototyped, simulated and synthesized using 65nm TSMC CMOS technology. At 720 MHz clock frequency, the proposed architecture processes 2K (1920×1080) resolution at 30 fps with ±27 (55×55) pixel search range using full search algorithm. Moreover, the proposed architecture is a flexible one and it can be used with different search algorithms to process higher resolutions such as 4K (3840×2160) resolution with 30 fps rate. To the best of our knowledge, the proposed architecture is one of the first ASIC motion estimation architectures in the literature for HEVC.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信