Towards Compact Single Image Super-Resolution via Contrastive Self-distillation

Wang, Yanbo; Lin, Shaohui; Qu, Yanyun; Wu, Haiyan; Zhang, Zhizhong; Xie, Yuan; Yao, Angela

doi:10.24963/ijcai.2021/155

Towards Compact Single Image Super-Resolution via Contrastive Self-distillation

Yanbo Wang, Shaohui Lin, Yanyun Qu, Haiyan Wu, Zhizhong Zhang, Yuan Xie, Angela Yao

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 1122-1128.

https://doi.org/10.24963/ijcai.2021/155

Keywords

Computer Vision: 2D and 3D Computer Vision

Abstract

Convolutional neural networks (CNNs) are highly successful for super-resolution (SR) but often require sophisticated architectures with heavy memory cost and computational overhead significantly restricts their practical deployments on resource-limited devices. In this paper, we proposed a novel contrastive self-distillation (CSD) framework to simultaneously compress and accelerate various off-the-shelf SR models. In particular, a channel-splitting super-resolution network can first be constructed from a target teacher network as a compact student network. Then, we propose a novel contrastive loss to improve the quality of SR images and PSNR/SSIM via explicit knowledge transfer. Extensive experiments demonstrate that the proposed CSD scheme effectively compresses and accelerates several standard SR models such as EDSR, RCAN and CARN. Code is available at https://github.com/Booooooooooo/CSD.

Yanbo Wang, Shaohui Lin, Yanyun Qu, Haiyan Wu, Zhizhong Zhang, Yuan Xie, and Angela Yao. 2021. Towards Compact Single Image Super-Resolution via Contrastive Self-distillation. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 1122–1128. https://doi.org/10.24963/ijcai.2021/155

@inproceedings{ijcai2021p155,
  title     = {Towards Compact Single Image Super-Resolution via Contrastive Self-distillation},
  author    = {Wang, Yanbo and Lin, Shaohui and Qu, Yanyun and Wu, Haiyan and Zhang, Zhizhong and Xie, Yuan and Yao, Angela},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {1122--1128},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/155},
  url       = {https://doi.org/10.24963/ijcai.2021/155},
}

TY  - CPAPER
TI  - Towards Compact Single Image Super-Resolution via Contrastive Self-distillation
AU  - Wang, Yanbo
AU  - Lin, Shaohui
AU  - Qu, Yanyun
AU  - Wu, Haiyan
AU  - Zhang, Zhizhong
AU  - Xie, Yuan
AU  - Yao, Angela
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 1122
EP  - 1128
DO  - 10.24963/ijcai.2021/155
UR  - https://doi.org/10.24963/ijcai.2021/155
ER  -