华南理工大学学报(自然科学版) ›› 2022, Vol. 50 ›› Issue (6): 80-90.doi: 10.12141/j.issn.1000-565X.210507

所属专题: 2022年电子、通信与自动控制

• 电子、通信与自动控制 • 上一篇    下一篇

特征域多假设预测视频压缩感知重构神经网络

杨春玲 凌茜 吕泽宇   

  1. 华南理工大学 电子与信息学院,广东 广州 510640
  • 收稿日期:2021-08-12 修回日期:2021-12-02 出版日期:2022-06-25 发布日期:2021-12-31
  • 通信作者: 杨春玲 (1970-),女,教授,主要从事图像/视频压缩编码、图像质量评价、图像/视频压缩感知重构研究。 E-mail:eeclyang@ scut. edu. cn
  • 作者简介:杨春玲 (1970-),女,教授,主要从事图像/视频压缩编码、图像质量评价、图像/视频压缩感知重构研究。
  • 基金资助:
    广东省自然科学基金资助项目 (2017A030311028,2019A1515011949)

Feature-domain Multi-Hypothesis Prediction Reconstruction Neural Network for Compressed Video Sensing

YANG Chunling  LING Xi  LÜ Zeyu#br#   

  1. School of Electronic and Information Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China
  • Received:2021-08-12 Revised:2021-12-02 Online:2022-06-25 Published:2021-12-31
  • Contact: 杨春玲 (1970-),女,教授,主要从事图像/视频压缩编码、图像质量评价、图像/视频压缩感知重构研究。 E-mail:eeclyang@ scut. edu. cn
  • About author:杨春玲 (1970-),女,教授,主要从事图像/视频压缩编码、图像质量评价、图像/视频压缩感知重构研究。
  • Supported by:
    Supported by the Natural Science Foundation of Guangdong Province (2017A030311028,2019A1515011949)

摘要: 在预测-残差重构框架中,利用视频时间相关性进行多假设预测是视频压缩感知重构的关键步骤。本文基于深度学习,研究了利用丰富的特征域信息实现更精确预测的方法,提出一种特征域多假设预测视频压缩感知重构网络(FMH_CVSNet)。首先提出一种新的特征域多假设预测模块(FMH_Module),通过构造合理的运动估计模块与假设权重求解模块增强了网络的预测能力;其次提出两阶段多参考帧运动补偿模式,使不同运动特征序列均能构造更优假设集,进一步提升了预测精度。仿真结果表明,FMH_CVSNet在各实验条件下均取得了优秀的重构性能,相比于传统多假设算法2sMHR平均PSNR提升了4.76dB,相比于基于深度学习的视频压缩感知重构算法VCSNet-2提升了3.87dB。

关键词: 视频压缩感知, 深度学习, 多假设预测, 自适应假设权重, 多参考帧, 视频运动特征

Abstract: In the prediction-residual reconstruction framework, multi-hypothesis prediction based on temporal correlation is the key step of compressed video sensing reconstruction. This paper studies the accuracy prediction method by utilizing rich features based on deep learning, and a novel feature-domain multi-hypothesis reconstruction network for compressed video sensing (FMH_CVSNet) is proposed. In FMH_CVSNet, the feature domain multi-hypothesis prediction module (FMH_Module) is firstly proposed, which improves the prediction ability by reasonably constructing the motion estimation module and the hypothesis weight calculation module based on the characteristics of video signal. Secondly, the two-stage multi-reference motion compensation mode is proposed, which makes the constructed hypothesis sets much better for sequences with different motion and the further improves the prediction accuracy. Simulation results show that FMH_CVSNet achieves better reconstruction performance under various experimental conditions, improves by 4.76dB compared with the traditional multi-hypothesis algorithm 2sMHR and improves by 3.87dB compared with CNN based compressed video sensing reconstruction algorithm VCSNet-2.

Key words: compressed video sensing, deep learning, multi-hypothesis prediction, adaptive hypothesis weight, multiple reference frame, video motion feature

中图分类号: