计算机应用 ›› 2013, Vol. 33 ›› Issue (12): 3540-3543.

• 多媒体处理技术 • 上一篇    下一篇

基于GPU加速的实时视频超分辨率重建

陈湘骥1,2,韩国强1,张芝源2   

  1. 1. 华南理工大学 计算机科学与工程学院,广州 510006
    2. 华南农业大学 信息学院,广州 510642;
  • 收稿日期:2013-06-13 修回日期:2013-08-16 出版日期:2013-12-01 发布日期:2013-12-31
  • 通讯作者: 陈湘骥
  • 作者简介: 陈湘骥(1976-),男,广西桂林人,讲师,博士研究生,CCF会员,主要研究方向:图像与视频恢复及超分辨率重建;
    韩国强(1962-), 男,江西临川人,教授,博士,主要研究方向:多媒体、图形图像处理、数字家庭;
    张芝源(1992-),男,广东惠州人,主要研究方向:视频超分辨率重建。
  • 基金资助:
    国家自然科学基金资助项目;国家自然科学广东联合基金项目;广东省工业科技计划项目

Real-time video super-resolution restruction based on GPU acceleration

CHEN Xiangji1,2,HAN Guoqiang2,ZHANG Zhiyuan1   

  1. 1. College of Information, South China Agricultural University, Guangzhou Guangdong 510642, China
    2. School of Computer Science and Engineering, South China University of Technology, Guangzhou Guangdong 510006, China
  • Received:2013-06-13 Revised:2013-08-16 Online:2013-12-31 Published:2013-12-01
  • Contact: CHEN Xiangji

摘要: 基于稀疏表示的超分辨率算法的图像重建质量好,但算法复杂,现有的CPU串行执行算法无法满足视频实时处理的需要。为此提出了基于GPU加速的稀疏表示的实时视频超分辨率算法。该算法着重于优化数据并行处理流程,提高GPU资源利用率,通过设置视频帧队列、提高显存访问并发率、采用主成分分析(PCA)降维、优化字典查找等手段,使算法执行速度比现有CPU串行算法提高了2个数量级,在显示分辨率为669×546的视频回放测试中达到每秒33帧。

关键词: 视频, 实时, 超分辨率, 稀疏表示, 通用计算图形处理器

Abstract: The methods of image super-resolution via sparse representation achieve good quality image reconstruction, but the CPU-based implementation of the methods hardly satisfies the requirement of real-time video super-resolution because of high computational complexity. Then, the method of real-time video super-resolution via sparse representation based on GPU acceleration was proposed. It focused on optimizing data parallel processing and improving resource utilization of GPU, including utilizing queues for video sequences, improving memory concurrent access rates, employing Principal Component Analysis (PCA) dimensionality reduction and optimizing dictionary querying operation. As a result, compared with the CPU-based implementation, the speed of data processing is increased two orders of magnitude, and the speed of playing a video with the size of 669×546 reaches 33 frames per second.

Key words: video, real-time, Super-Resolution (SR), sparse representation, General Purpose Graphic Processing Unit (GPGPU)

中图分类号: