电子学报 ›› 2021, Vol. 49 ›› Issue (3): 435-442.DOI: 10.12263/DZXB.20200272

所属专题: 压缩感知

• 学术论文 • 上一篇    下一篇

基于帧间组稀疏的两阶段递归增强视频压缩感知重构网络

禤韵怡, 杨春玲   

  1. 华南理工大学电子与信息学院, 广东广州 510640
  • 收稿日期:2020-03-17 修回日期:2020-08-20 出版日期:2021-03-25 发布日期:2021-03-25
  • 通讯作者: 杨春玲
  • 作者简介:禤韵怡 女,1995年生于广东佛山,华南理工大学电子与信息学院研究生.研究方向:视频压缩感知.E-mail:eeyyxuan@mail.scut.edu.cn
  • 基金资助:
    广东省自然科学基金-重点(No.2017A030311028);广东省自然科学基金(No.2019A1515011949)

Two-Stage Recursive Enhancement Reconstruction Based on Video Inter-frame Group Sparse Representation in Compressed Video Sensing

XUAN Yun-yi, YANG Chun-ling   

  1. School of Electronic and Information Engineering, South China University of Technology, Guangzhou, Guangdong 510640, China
  • Received:2020-03-17 Revised:2020-08-20 Online:2021-03-25 Published:2021-03-25

摘要: 基于迭代优化的传统视频压缩感知重构算法运行时间长,参数的自适应性较低,限制了其实用性和泛化能力.利用神经网络强大的计算能力和运行速度快、参数可学习的优点,本文首先提出了帧间组稀疏网络(VGSR-Net),用神经网络将图像块组映射到更高维的稀疏表示域中,并利用可学习的阈值提取帧间相关特征.在此基础上,提出了两阶段混合递归增强重构网络(2sRER-VGSR-Net).首先,利用VGSR-Net对初始重构结果进行初步增强;然后,引入STMC-Net实现运动估计,并利用残差重构网络进一步重构当前帧丢失的信息,得到更高质量的重构结果.在第二阶段重构过程中采用混合递归结构,充分利用已有的高质量重构帧信息.仿真结果表明,所提算法与现有最优迭代优化重构算法SSIM-InterF-GSR相比重构性能提升了1.99dB;和基于深度学习的重构网络CSVideoNet相比,性能提升了4.60dB.

 

关键词: 视频压缩感知, 深度学习, 帧间组稀疏表示, 混合递归网络, 运动估计, 增强重构

Abstract: The traditional iterative optimized based video compression sensing algorithms are limited by long running time and low adaptability of parameters,resulting in low practicability and generalization.Taking advantage of the powerful computing power,fast speed and learnable parameters of neural networks,this paper first proposes a group sparse representation network (VGSR-Net),which maps the image block group to a higher-dimensional sparse domain through convolution,and uses a learnable threshold to denoise and extract inter-frame correlation.On this basis,a two-stage recursive enhance reconstruction network(2sRER-VGSR-Net) is proposed.First,we perform VGSR-Net to preliminarily enhance the initial reconstruction and then introduce STMC-Net as motion estimation,and the compensated frames are fed into the residual reconstruction network to further extract the missing detail and enhance the current frame.The second stage of reconstruction adopts a hybrid recursive structure with the aim of making full use of the existing better quality reconstructed frames.The simulation results show that the proposed algorithm improves the PSNR(Reak Signal to Noise Ratio) by 1.99dB compared with the existing state-of-art traditional compressed video sensing reconstruction algorithms SSIM-InterF-GSR,while improves the PSNR by 4.60dB with the comparation of the network-based algorithm CSVideoNet.

Key words: compressed video sensing, deep learning, group based sparse representation, hybrid recursive network, motion estimation, enhancement reconstruction

中图分类号: