中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Semantic-Aware Visual Decomposition for Image Coding

文献类型：期刊论文

作者Chang, Jianhui3; Zhang, Jian4; Li, Jiguo1; Wang, Shiqi5; Mao, Qi2; Jia, Chuanmin3; Ma, Siwei3; Gao, Wen3
刊名INTERNATIONAL JOURNAL OF COMPUTER VISION
出版日期2023-06-02
页码23
关键词Image coding Semantic-aware visual decomposition Structure-texture Coherency regularization Extremely low bitrate
ISSN号0920-5691
DOI10.1007/s11263-023-01809-7
英文摘要In this paper, we propose a novel image coding framework with semantic-aware visual decomposition towards extremely low bitrate compression. In particular, an input image is analyzed into a semantic map as structural representation and semantic-wise texture representation and further compressed into bitstreams at the encoder side. On the decoder side, the received bitstreams of dual-layer representations are decoded and reconstructed for target image synthesis with generative models. Moreover, the attention mechanism is introduced into the model architecture for texture representation modeling and a coherency regularization is proposed to further optimize the texture representation space by aligning the representation space with the source pixel space for higher synthesis quality. Besides, we also propose a cross-channel entropy module and control the quantization scale to facilitate rate-distortion optimization. Upon compressing the decomposed components into the bitstream, the simple yet effective representation philosophy benefits image compression in many aspects. First, in terms of compression performance, compact representations, and high visual synthesis quality can bring remarkable advantages. Second, the proposed framework yields a physically explainable bitstream composed of the structural segment and semantic-wise texture segments. Third and most importantly, subsequent vision tasks (e.g., content manipulation) can receive fundamental support from the semantic-aware visual decomposition and synthesis mechanism. Extensive experimental results demonstrate the superiority of the proposed framework towards efficient visual representation learning, high efficiency image compression (< 0.1 bpp), and intelligent visual applications (e.g., manipulation and analysis).
WOS研究方向Computer Science
语种英语
WOS记录号WOS:001000503000001
出版者SPRINGER
源URL[http://119.78.100.204/handle/2XEOYT63/21463]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Zhang, Jian; Ma, Siwei
作者单位1.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
2.Commun Univ China, State Key Lab Media Convergence & Commun, Beijing 100024, Peoples R China
3.Peking Univ, Natl Engn Res Ctr Visual Technol, Sch Comp Sci, Beijing 100871, Peoples R China
4.Peking Univ, Sch Elect & Comp Engn, Shenzhen Grad Sch, Shenzhen 518055, Peoples R China
5.City Univ Hong Kong, Dept Comp Sci, Hong Kong, Peoples R China
推荐引用方式
GB/T 7714		 Chang, Jianhui,Zhang, Jian,Li, Jiguo,et al. Semantic-Aware Visual Decomposition for Image Coding[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2023:23.
APA Chang, Jianhui.,Zhang, Jian.,Li, Jiguo.,Wang, Shiqi.,Mao, Qi.,...&Gao, Wen.(2023).Semantic-Aware Visual Decomposition for Image Coding.INTERNATIONAL JOURNAL OF COMPUTER VISION,23.
MLA Chang, Jianhui,et al."Semantic-Aware Visual Decomposition for Image Coding".INTERNATIONAL JOURNAL OF COMPUTER VISION (2023):23.

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。