基于深度残差网络的HEVC压缩视频增强  

Deep Residual Network Based HEVC Compressed Videos Enhancement

作　　者：何晓艺 段凌宇 林巍峣[1] HE Xiao-yi;DUAN Ling-yu;LIN Wei-yao(School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;National Engineering Laboratory for Video Technology,Peking University,Beijing 100871,China)

机构地区：[1]上海交通大学电子信息与电气工程学院,上海200240 [2]北京大学数字视频编解码技术国家工程实验室,北京100871

出　　处：《计算机科学》2019年第3期88-91,共4页Computer Science

基　　金：国家自然科学基金项目(61471235);上海市"一带一路"青年科学家交流国际合作项目(17510740100);Ng Teng Fong慈善基金PKU-NTU联合研究中心项目(JRI)资助

摘　　要：文中提出了一种基于深度残差网络的HEVC压缩视频增强方法。该方法利用一系列级联的残差模块来完成特征提取,然后基于这些特征进行视频的质量增强。与现有的方法相比,所提方法能够捕捉到压缩视频帧更清晰和泛化的特征。实验结果表明,所提方法在20个通用的测试视频序列上能够实现平均6.92%的BD-rate增益,是所有参与比较的方法中效果最好的。This paper proposed a HEVC-compressed videos enhancement method based on deep residual network.This method utilizes several stacked residual blocks to achieve feature extraction,followed by feature enhancement and reconstruction.Compared with the existing methods which only use a few convolutional layers,the proposed method can capture the feature of input compressed frames in a more distinctive and stable way.Experimental results show that the proposed method leads to over 6.92% BD-rate saving on 20 benchmark sequences and achieves the best performance among the compared methods.

关 键 词：高效率视频编码 卷积神经网络 残差网络 压缩视频增强 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]