可重构结构下合成视点失真变化算法并行设计与实现

doi:10.11772/j.issn.1001-9081.2020091462

计算机应用 ›› 2021, Vol. 41 ›› Issue (6): 1734-1740.DOI: 10.11772/j.issn.1001-9081.2020091462

所属专题：先进计算

可重构结构下合成视点失真变化算法并行设计与实现

蒋林, 施佳琪, 李远成

西安科技大学计算机科学与技术学院, 西安 710600

收稿日期:2020-09-21 修回日期:2020-12-03 出版日期:2021-06-10 发布日期:2020-12-28
通讯作者: 施佳琪
作者简介:蒋林(1970-),男,陕西杨凌人,教授,博士,主要研究方向:专用集成电路设计、计算机体系结构、计算机图形图像处理;施佳琪(1994-),男,陕西西安人,硕士研究生,主要研究方向:计算机体系结构;李远成(1981-),男,河南开封人,讲师,博士,CCF会员,主要研究方向:计算机体系结构、并行计算、人工智能。
基金资助:
国家自然科学基金重点项目（61834005）；陕西省自然科学基金面上项目（2020JM-525）。

Parallel design and implementation of synthetic view distortion change algorithm in reconfigurable structure

JIANG Lin, SHI Jiaqi, LI Yuancheng

College of Computer Science and Technology, Xi'an University of Science and Technology, Xi'an Shaanxi 710600, China

Received:2020-09-21 Revised:2020-12-03 Online:2021-06-10 Published:2020-12-28
Supported by:
This work is partially supported by the Key Project of National Natural Science Foundation of China (61834005), the General Project of Natural Science Foundation of Shaanxi Province (2020JM-525).

摘要/Abstract

摘要： 针对三维高效视频编码（3D-HEVC）中，基于深度图的合成视点失真变化（SVDC）算法存在计算时间复杂度较高的问题，提出了一种在可重构阵列结构下基于混合粒度的SVDC算法并行化方法。首先，将SVDC算法分为虚拟视点合成（VVS）和失真值计算两个部分。其次，VVS部分采用流水线作业方式加速，而失真值计算部分采用两级划分加速：任务级——将合成后的图像按照像素点进行划分，指令级——将像素点内部的失真值按照计算过程进行划分。最后，采用可重构机制将VVS部分和失真值计算部分进行并行化处理。理论分析和硬件仿真结果表明，在执行时间上，采用4个处理单元（PE）的该方法具有2.11的加速比性能，与基于底层虚拟机（LLVM）和共享存储并行编程（OpenMP）的SVDC算法相比，计算时间分别缩短了18.56%和21.93%。可见所提方法能挖掘SVDC算法的并行性，并结合可重构阵列结构特点有效缩短了SVDC算法的执行时间。

关键词: 合成视点失真变化算法, 高效视频编码, 可重构结构, 混合粒度, 并行化映射

Abstract: Focused on the high computational time complexity of the depth map based Synthesized View Distortion Change (SVDC) algorithm in 3D High Efficiency Video Coding (3D-HEVC), a new parallelization method of SVDC algorithm based on hybrid granularity was proposed under the reconfigurable array structure. Firstly, the SVDC algorithm was divided into two parts:Virtual View Synthesis (VVS) and distortion value calculation. Secondly, the VVS part was accelerated by pipeline operation, and the distortion value calculation part was accelerated by dividing into two levels:task level, which means dividing the synthesized image according to pixels, and instruction level, that is dividing the distortion values inside the pixel by the calculation process. Finally, a reconfigurable mechanism was used to parallelize the VVS part and distortion value calculation part. Theoretical analysis and hardware simulation results show that, in terms of execution time, the proposed method has the speedup ratio of 2.11 with 4 Process Elements (PEs). Compared with the SVDC algorithms based on Low Level Virtual Machine (LLVM) and Open Multi-Processing (OpenMP), the proposed method has the calculation time reduced by 18.56% and 21.93% respectively. It can be seen that the proposed method can mine the parallelism of the SVDC algorithm, and effectively shorten the execution time of the SVDC algorithm by combining with the characteristics of the reconfigurable array structure.

Key words: Synthesized View Distortion Change (SVDC) algorithm, High Efficiency Video Coding (HEVC), reconfigurable structure, hybrid granularity, parallelized mapping

中图分类号:

TP303

蒋林, 施佳琪, 李远成. 可重构结构下合成视点失真变化算法并行设计与实现[J]. 计算机应用, 2021, 41(6): 1734-1740.

JIANG Lin, SHI Jiaqi, LI Yuancheng. Parallel design and implementation of synthetic view distortion change algorithm in reconfigurable structure[J]. Journal of Computer Applications, 2021, 41(6): 1734-1740.

参考文献

[1] 贾克斌, 刘畅, 刘鹏宇, 等. 基于3D-HEVC的三维视频编码方法综述[J]. 北京工业大学学报, 2019, 45(11):1082-1094.(JIA K B,LIU C,LIU P Y,et al. Three-dimensional video coding based on 3D-HEVC:a survey[J]. Journal of Beijing University of Technology,2019,45(11):1082-1094.)
[2] LIAO Y W,CHEN M J,YEH C H,et al. Efficient inter-prediction depth coding algorithm based on depth map segmentation for 3DHEVC[J]. Multimedia Tools and Applications,2019,78(8):10181-10205.
[3] ZHANG Y,ZHANG H,YU M,et al. Sparse representation-based video quality assessment for synthesized 3D videos[J]. IEEE Transactions on Image Processing,2020,29:509-524.
[4] 贾克斌, 刘畅, 窦环. 基于自适应分段跳过规则的视点合成优化方法[J]. 高技术通讯, 2017, 27(9/10):777-782.(JIA K B,LIU C,DOU H. Optimization of viewpoint synthesis based on adaptive segment skip rule[J]. Chinese High Technology Letters,2017,27(9/10):777-782.)
[5] 蒋林, 王杏军, 刘镇弢, 等. 基于SystemC的可重构阵列处理器模型[J]. 西安邮电大学学报, 2016, 21(3):73-78.(JIANG L, WANG X J,LIU Z T,et al. Reconfigurable array processor modeling based on SystemC[J]. Journal of Xi' an University of Posts and Telecommunications,2016,21(3):73-78.)
[6] 袁开坚, 张兴明, 高彦钊. 基于并行度最大化的多目标优化任务划分算法[J]. 计算机应用, 2017, 37(7):1916-1920.(YUAN K J, ZHANG X M,GAO Y Z. Task partitioning algorithm based on parallelism maximization with multi-objective optimization[J]. Journal of Computer Applications,2017,37(7):1916-1920.)
[7] ZAHAF H E,LIPARI G,BERTOGNA M,et al. The parallel multimode digraph task model for energy-aware real-time heterogeneous multi-core systems[J]. IEEE Transactions on Computers,2019, 68(10):1511-1524.
[8] BONIFACI V,WIESE A,BARUSH S K,et al. A generalized parallel task model for recurrent real-time processes[J]. ACM Transactions on Parallel Computing,2019,6(1):Article No. 3.
[9] XIAO B,WANG H,WU J,et al. A multi-grained parallel solution for HEVC encoding on heterogeneous platforms[J]. IEEE Transactions on Multimedia,2019,21(12):2997-3009.
[10] SANCHEZ G,SALDANHA M,AGOSTINI L,et al. Analysis of parallel encoding using tiles in 3D high efficiency video coding[J]. Signal,Image and Video Processing,2019,13(6):1079-1086.
[11] LIM K, RO W W. Contents-aware partitioning algorithm for parallel high efficiency video coding[J]. Multimedia Tools and Applications,2019,78(9):11427-11442.
[12] JIANG L,LIU Y,SHAN R,et al. RDMM:runtime dynamic migration mechanism of distributed cache for reconfigurable array processor[J]. Integration,2020,72:82-91.
[13] LIU W,YANG L,JIANG W,et al. Thermal-aware task mapping on dynamically reconfigurable Network-on-Chip based multiprocessor System-on-Chip[J]. IEEE Transactions on Computers,2018,67(12):1818-1834.
[14] AHN Y J,YOO J,JO H H,et al. Software pipelining with CGA and proposed intrinsics on a reconfigurable processor for HEVC decoders[J]. Journal of Real-Time Image Processing,2019,16(6):2173-2187.
[15] HUANG J C, HUANG H C, LIU H H. Research on the parallelization of image quality analysis algorithm based on deep learning[J]. Journal of Visual Communication and Image Representation,2020,71:Article No. 102709.

可重构结构下合成视点失真变化算法并行设计与实现

Parallel design and implementation of synthetic view distortion change algorithm in reconfigurable structure

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

编辑推荐

Metrics

[1]	周怡钊, 王晓东, 章联军, 兰琼琼. 基于Logistic和Arnold变换的HEVC选择性加密方案[J]. 计算机应用, 2019, 39(10): 2973-2979.
[2]	郭磊, 王晓东, 徐博文, 王健. 基于HEVC的帧内预测模式决策和编码单元划分快速算法[J]. 计算机应用, 2018, 38(4): 1157-1163.
[3]	蔡春亭, 冯桂, 王驰, 韩雪. 基于帧内预测模式多划分的HEVC鲁棒视频水印算法[J]. 计算机应用, 2017, 37(6): 1772-1776.
[4]	陈甬娜, 周宇, 王晓东, 郭磊. 基于菱形编码的视频信息隐藏算法[J]. 计算机应用, 2017, 37(10): 2806-2812.
[5]	赵利平, 林涛, 龚迅炜, 朱蓉. 帧内微块复制的屏幕图像编码算法[J]. 计算机应用, 2016, 36(7): 1938-1943.
[6]	陈智贤, 王国中, 赵海武, 李国平, 滕国伟. 改进的基于AVS2的样点自适应补偿滤波算法[J]. 计算机应用, 2016, 36(5): 1362-1365.
[7]	李鹏, 彭宗举, 李持航, 陈芬. 基于纹理特性与空域相关的高效视频编码帧内分级快速算法[J]. 计算机应用, 2016, 36(4): 1085-1091.
[8]	刘颖, 高雪明, 林庆帆. 高效视频编码帧内快速深度决策算法[J]. 计算机应用, 2016, 36(10): 2854-2858.
[9]	张明辉, 冯桂. 基于纹理方向的高效视频编码鲁棒视频水印[J]. 计算机应用, 2015, 35(9): 2531-2534.
[10]	陈先义, 赵利平, 陈治中, 林涛. 二维帧内串匹配屏幕图像编码算法[J]. 计算机应用, 2015, 35(9): 2640-2647.