HEVC运动估计中SAD算法的动态可重构实现
|
摘要
高效视频编码(HEVC)标准中引入的不对称分割模式导致运动估计算法中绝对差值和(SAD)运算量成倍增加.为了提高运动估计算法的执行效率,方便用户进行自主选择,设计了同时支持不对称分割模式开启和关闭2种执行模式以及执行模式间自由切换的可重构阵列结构.为了满足用户要求编码速度的同时,最大限度地利用可重构阵列处理器的资源,在阵列结构为16×16个处理元中通过加载16×8、16×4以及16×2个处理元的指令来进行阵列规模的动态重构,采用指令下发的方式将不同的指令发送到对应处理元进行相应配置.实验结果表明,所提出的可重构实现方式在硬件资源占用量接近条件下,相较于流水化实现处理时间减少了约35%,吞吐量提高了约0.4倍.该实现具有较高的执行效率,能够进行执行模式与阵列规模的切换,具有较好的灵活性.
The asymmetric partitioning mode introduced in the high efficiency video coding( HEVC)standard results in a double increase of the sum of absolute difference( SAD) operation amount in the motion estimation algorithm. In order to improve the efficiency of motion estimation algorithm,it is convenient for users to choose independently,a reconfigurable array structure is designed which supports both the opening and closing of an asymmetric partitioning mode and the free switching between execution modes. In order to satisfy the user's requirement for coding speed,and maximize the use of the resources of the reconfigurable array processor,16 × 8,16 × 4,and 16 × 2 processing elements are loaded in an array structure of 16 × 16 processing elements. The instruction is used to dynamically reconfigure the array size,and different instructions are sent to corresponding processing elements for corresponding configuration by means of the instruction issuance manner. The experimental results show that the proposed reconfigurable implementation approach reduces the processing time by about 35% and the throughput is improved by about 0. 4 times compared with the streamlining under the condition that the hardware resourceoccupancy closely. The implementation has high execution efficiency,and can switch between execution mode and array size,and has better flexibility.
The asymmetric partitioning mode introduced in the high efficiency video coding( HEVC)standard results in a double increase of the sum of absolute difference( SAD) operation amount in the motion estimation algorithm. In order to improve the efficiency of motion estimation algorithm,it is convenient for users to choose independently,a reconfigurable array structure is designed which supports both the opening and closing of an asymmetric partitioning mode and the free switching between execution modes. In order to satisfy the user's requirement for coding speed,and maximize the use of the resources of the reconfigurable array processor,16 × 8,16 × 4,and 16 × 2 processing elements are loaded in an array structure of 16 × 16 processing elements. The instruction is used to dynamically reconfigure the array size,and different instructions are sent to corresponding processing elements for corresponding configuration by means of the instruction issuance manner. The experimental results show that the proposed reconfigurable implementation approach reduces the processing time by about 35% and the throughput is improved by about 0. 4 times compared with the streamlining under the condition that the hardware resourceoccupancy closely. The implementation has high execution efficiency,and can switch between execution mode and array size,and has better flexibility.
引文
[1]Kibeya H,Belghith F,Loukil H,et al.TZSearch pattern search improvement for HEVC motion estimation modules[M].2014 1stInternational Conference on Advanced Technologies for Signal and Image Processing(ATSIP),Sousse,2014:95-99.
[2]黄卫锋,桑红石,郑兆青,等.用于H.264的高性能整像素运动估计VLSI的设计[J].微电子学,2007,37(2):260-264.
[3]Tung D M,Dong T L T.A VLSI architecture for H.264/AVC variable block size motion estimation[J].Journal of Automation&Control Engineering,2015,3(1):51-55.
[4]Grellert,M,Sampaio,F,et al.A multilevel data reuse scheme for motion estimation and its VLSI design[C]∥ISCAS.Rio de Janeiro:Circuits and Systems(ISCAS),2011 IEEE International Symposium of Circuits and Systems(ISCAS),Rio de Janeiro,2011,583-586.
[5]Mariusz Jakubowski.Optimization of the adaptive computationally-scalable motion estimation and compensation for the hardware H.264/AVC Encoder[J].Journal of Signal Processing Systems,2016,82(3):391-402.
[6]Chao-Yang Kao.A memory-efficient and highly parallel architecture for variable block size integer motion estimation in H.264/AVC[J].IEEE Transactions on Very Large Scale Integration Systems,2010,18(6):866-874.
[7]Byun J,Jung Y,Kim J.Design of integer motion estimator of HEVC for asymmetric motion-partitioning mode and4K-UHD[J].Electronics Letters,2013,49(18):1142-1143.
[8]Alcocer E,Gutierrez R,Lopez-Granado O,et al.Design and implementation of an efficient hardware integer motion estimator for an HEVC video encoder[J].Journal of Real-Time Image Processing,2016,11(1):1-11.
[9]李涛,蒋林,刘镇弢,等.一种新型阵列视频信号处理单元结构[P],中国,(中华人民共和国国家产权局)20012中国公开,201110046537.8.
[10]韩佳.基于可重构系统的HEVC运动估计算法实现[J].信息技术,2016,40(11):12-17.
[11]Gary J.Sullivan.Overview of the high efficiency video coding(HEVC)standard[J].IEEE Transactions on Circuits&Systems for Video Technology,2012,22(12):1649-1668.
[12]Il-Koo Kim.Block partitioning structure in the HEVC standard[J].IEEE Transactions on Circuits&Systems for Video Technology,2012,22(12):1697-1706.
[2]黄卫锋,桑红石,郑兆青,等.用于H.264的高性能整像素运动估计VLSI的设计[J].微电子学,2007,37(2):260-264.
[3]Tung D M,Dong T L T.A VLSI architecture for H.264/AVC variable block size motion estimation[J].Journal of Automation&Control Engineering,2015,3(1):51-55.
[4]Grellert,M,Sampaio,F,et al.A multilevel data reuse scheme for motion estimation and its VLSI design[C]∥ISCAS.Rio de Janeiro:Circuits and Systems(ISCAS),2011 IEEE International Symposium of Circuits and Systems(ISCAS),Rio de Janeiro,2011,583-586.
[5]Mariusz Jakubowski.Optimization of the adaptive computationally-scalable motion estimation and compensation for the hardware H.264/AVC Encoder[J].Journal of Signal Processing Systems,2016,82(3):391-402.
[6]Chao-Yang Kao.A memory-efficient and highly parallel architecture for variable block size integer motion estimation in H.264/AVC[J].IEEE Transactions on Very Large Scale Integration Systems,2010,18(6):866-874.
[7]Byun J,Jung Y,Kim J.Design of integer motion estimator of HEVC for asymmetric motion-partitioning mode and4K-UHD[J].Electronics Letters,2013,49(18):1142-1143.
[8]Alcocer E,Gutierrez R,Lopez-Granado O,et al.Design and implementation of an efficient hardware integer motion estimator for an HEVC video encoder[J].Journal of Real-Time Image Processing,2016,11(1):1-11.
[9]李涛,蒋林,刘镇弢,等.一种新型阵列视频信号处理单元结构[P],中国,(中华人民共和国国家产权局)20012中国公开,201110046537.8.
[10]韩佳.基于可重构系统的HEVC运动估计算法实现[J].信息技术,2016,40(11):12-17.
[11]Gary J.Sullivan.Overview of the high efficiency video coding(HEVC)standard[J].IEEE Transactions on Circuits&Systems for Video Technology,2012,22(12):1649-1668.
[12]Il-Koo Kim.Block partitioning structure in the HEVC standard[J].IEEE Transactions on Circuits&Systems for Video Technology,2012,22(12):1697-1706.