跳到主要导航 跳到搜索 跳到主要内容

AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring

  • East China Normal University

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

摘要

Despite the recent progress in enhancing the efficacy of image deblurring, the limited decoding capability constrains the upper limit of State-Of- The-Art (SOTA) methods. This paper proposes a pioneering work, Adaptive Patch Ex-iting Reversible Decoder (AdaRevD), to explore their in-sufficient decoding capability. By inheriting the weights of the well-trained encoder, we refactor a reversible de-coder which scales up the single-decoder training to multi-decoder training while remaining GPU memory-friendly. Meanwhile, we show that our reversible structure gradually disentangles high-level degradation degree and low-level blur pattern (residual of the blur image and its sharp counterpart) from compact degradation representation. Besides, due to the spatially-variant motion blur kernels, different blur patches have various deblurring difficulties. We further introduce a classifier to learn the degradation degree of image patches, enabling them to exit at different sub-decoders for speedup. Experiments show that our AdaRevD pushes the limit of image deblurring, e.g., achieving 34.60 dB in PSNR on GoPro dataset.

源语言英语
主期刊名Proceedings - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
出版商IEEE Computer Society
页25681-25690
页数10
ISBN(电子版)9798350353006
DOI
出版状态已出版 - 2024
活动2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024 - Seattle, 美国
期限: 16 6月 2024 → 22 6月 2024

丛书

姓名Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN(印刷版)1063-6919

会议

会议2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
国家/地区美国
市Seattle
时期16/06/24 → 22/06/24

学术指纹

探究 'AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring' 的科研主题。它们共同构成独一无二的学术指纹。

引用此