Lightweight Multiattention Recursive Residual CNN-Based In-Loop Filter Driven by Neuron Diversity
文献类型：期刊论文
| 作者 | Li, Mingxuan2,3; Ji, Wen1,3 |
| 刊名 | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
 |
| 出版日期 | 2023-11-01 |
| 卷号 | 33期号:11页码:6996-7008 |
| 关键词 | Video coding in-loop filtering convolutional neural network deep learning high-efficiency video coding (HEVC) |
| ISSN号 | 1051-8215 |
| DOI | 10.1109/TCSVT.2023.3270729 |
| 英文摘要 | Many convolutional neural network (CNN)-based in-loop filters have been proposed to improve coding performance. However, considering the single perception scale, high parameter complexity, and the need to train multiple models for various quantization parameters (QPs), the performance and practicability of most existing methods are limited. Inspired by neuron diversity, this paper proposes a lightweight multiattention recursive residual CNN-based in-loop filter that can handle encoded frames with various QP values, frame types (FTs), and temporal layers (TLs) via a single model. First, multiscale features are learned in the neural network and fused with the proposed multidensity block (MDB) and multiscale fusion attention group (MFAG). Second, a recursive structure is adopted to improve the model depth while saving many parameters. The proposed auxiliary parameter fusion attention (APFA) and long-short-term skip connection (LSTSC) models integrate QPs, FTs and TLs into the model while accelerating training. Finally, we propose implementing LMA-RRCNN in parallel with the standard in-loop filter and select the optimal enhanced result in each patch. The experimental results on standard test sequences show that the proposed method achieves on average 13.70% and 11.87% BD rate savings under all-intra and random-access configurations, respectively, outperforming other state-of-the-art approaches. |
| 资助项目 | Beijing Natural Science Foundation[L221004] ; National Key Research and Development Program of China[2022YFF0902403] ; National Key Research and Development Program of China[2022YFE0125400] ; National Natural Science Foundation of China[62072440] |
| WOS研究方向 | Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001093434100054 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/38107]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Ji, Wen |
| 作者单位 | 1.Peng Cheng Lab, Shenzhen 518055, Peoples R China 2.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100190, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China |
| 推荐引用方式 GB/T 7714 | Li, Mingxuan,Ji, Wen. Lightweight Multiattention Recursive Residual CNN-Based In-Loop Filter Driven by Neuron Diversity[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2023,33(11):6996-7008. |
| APA | Li, Mingxuan,&Ji, Wen.(2023).Lightweight Multiattention Recursive Residual CNN-Based In-Loop Filter Driven by Neuron Diversity.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,33(11),6996-7008. |
| MLA | Li, Mingxuan,et al."Lightweight Multiattention Recursive Residual CNN-Based In-Loop Filter Driven by Neuron Diversity".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 33.11(2023):6996-7008. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。