31st International Joint Conference on Artificial Intelligence (IJCAI) 2022

Lightweight Bimodal Network for Single-Image Super-Resolution
via Symmetric CNN and Recursive Transformer

Guangwei Gao1†     Zhengxue Wang1†     Juncheng Li2*     Wenjie Li1     Yi Yu3     Tieyong Zeng2    

1 Nanjing University of Posts and Telecommunications     2 The Chinese University of Hong Kong    
3 National Institute of Informatics    

  †Co-first authors, *Corresponding author      Contact us: {csggao,cvjunchengli}@gmail.com, wzx_0826@163.com


Single-image super-resolution (SISR) has achieved significant breakthroughs with the development of deep learning. However, these methods are difficult to be applied in real-world scenarios since they are inevitably accompanied by the problems of computational and memory costs caused by the complex operations. To solve this issue, we propose a Lightweight Bimodal Network (LBNet) for SISR. Specifically, an effective Symmetric CNN is designed for local feature extraction and coarse image reconstruction. Meanwhile, we propose a Recursive Transformer to fully learn the long-term dependence of images thus the global information can be fully used to further refine texture details. Studies show that the hybrid of CNN and Transformer can build a more efficient model. Extensive experiments have proved that our LBNet achieves more prominent performance than other state-of-the-art methods with a relatively low computational cost and memory consumption. The code is available at https://github.com/IVIPLab/LBNet.


The complete architecture of the proposed Lightweight Bimodal Network (LBNet).

The architecture of the proposed Local Feature Fusion Module (LFFM) and Feature Refinement Dual-Attention Block (FRDAB).


Visual Results


Paper : [ Paper ]
Supp Material : [ Supplementary Material(提取码:3ywm) ]
SR Results : [ SR Results(提取码:xpuh) ]
Source Code : [ Code ]


    title = {Lightweight Bimodal Network for Single-Image Super-Resolution via Symmetric CNN and Recursive Transformer},
    author = {Gao, Guangwei and Wang, Zheengxue and Li, Juncheng and Li, Wenjie and Yu, Yi and Zeng, Tieyong},
    booktitle = {IJCAI},
    year = {2022}