Distillation-Supervised Convolutional Low-Rank Adaptation for Efficient Image Super-Resolution

📖 The Architecture of DSCLoRA Model

We replace the SPAB module with the proposed SConvLB module and incorporate ConvLoRA layers into both the pixel shuffle block and its preceding convolutional layer. Spatial Affinity Distillation Loss is calculated between each feature map.

🚀 Updates

• [2025.06.11] 🎉 Our work is selected to give a oral presentation at the NTIRE 2025 workshop at CVPR.
• [2025.04.21] ✅ Upload our model on Hugging Face 🤗.
• [2025.04.15] 🎉 Our paper is accepted to CVPR 2025 Workshop!
• [2025.03.26] 🏆 Our team won 1st place in the NTIRE 2025 Efficient SR Challenge. Challenge report is here.
• [2025.03.21] ✅ Release our code here.

🔧 The Environments

The evaluation environments adopted by us is recorded in the requirements.txt. After you built your own basic Python (Python = 3.9 in our setting) setup via either virtual environment or anaconda, please try to keep similar to it via:

• Step1: install Pytorch first: pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu117

• Step2: install other libs via: pip install -r requirements.txt

or take it as a reference based on your original environments.

⚡ How to test the model?

1. Run the run.sh

   CUDA_VISIBLE_DEVICES=0 python test_demo.py --data_dir [path to your data dir] --save_dir [path to your save dir] --model_id 23

   • Be sure the change the directories --data_dir and --save_dir.

🥰 Citation

If our work is useful to you, please use the following BibTeX for citation.

@inproceedings{Chai2025DistillationSupervisedCL,
  title={Distillation-Supervised Convolutional Low-Rank Adaptation for Efficient Image Super-Resolution},
  author={Xinning Chai and Yao Zhang and Yuxuan Zhang and Zhengxue Cheng and Yingsheng Qin and Yucai Yang and Li Song},
  year={2025},
  url={https://api.semanticscholar.org/CorpusID:277787382}
}

📜 License and Acknowledgement

This code repository is release under MIT License.