This is the official implementation of our CVPR 2025 paper:
Boosting Domain Incremental Learning: Selecting the Optimal Parameters is All You Need
conda create -n soyo python=3.8
conda activate soyo
pip install -r requirements.txtPlease refer to DomainNet Project to download the dataset or run:
cd dil_dataset
bash download_domainnet.shThen unzip the downloaded files, and confirm the file directory as shown below:
DomainNet
├── clipart
│ ├── aircraft_carrier
│ ├── airplane
│ ... ...
├── clipart_test.txt
├── clipart_train.txt
├── infograph
│ ├── aircraft_carrier
│ ├── airplane
│ ... ...
├── infograph_test.txt
├── infograph_train.txt
├── painting
│ ├── aircraft_carrier
│ ├── airplane
... ...
Please refer to CDDB Project and download the dataset from CDDB Dataset.
Then unzip the downloaded files, and confirm the file directory as shown below:
CDDB
├── biggan
│ ├── train
│ └── val
├── gaugan
│ ├── train
│ └── val
├── san
│ ├── train
│ └── val
├── whichfaceisreal
│ ├── train
│ └── val
├── wild
│ ├── train
│ └── val
... ...
Please refer to CORe50 Project and download the file shown below:
CORe50
├── core50_imgs.npz
├── labels.pkl
├── LUP.pkl
└── paths.pkl
Please confirm the path of your datasets in the config files.
python main.py --config configs/domainnet_soyo_vit.yaml --device 0
python main.py --config configs/domainnet_soyo_clip.yaml --device 0
python main.py --config configs/cddb_soyo_vit.yaml --device 0
python main.py --config configs/cddb_soyo_clip.yaml --device 0
python main.py --config configs/core50_soyo_vit.yaml --device 0
python main.py --config configs/core50_soyo_clip.yaml --device 0
Please refer to LDB project.
Please refer to LNA paper.
We thank PyCIL and S-Prompts for their wonderful framework and codes!
We also thank CLIP and CoOp for their helpful components.
If any part of our paper and code is helpful to your research, please consider citing the following bib entry:
@inproceedings{wang2025boosting,
title={Boosting Domain Incremental Learning: Selecting the Optimal Parameters is All You Need},
author={Wang, Qiang and Song, Xiang and He, Yuhang and Han, Jizhou and Ding, Chenhao and Gao, Xinyuan and Gong, Yihong},
booktitle={Proceedings of the Computer Vision and Pattern Recognition Conference},
pages={4839--4849},
year={2025}
}
@inproceedings{wang2024non,
title={Non-exemplar domain incremental learning via cross-domain concept integration},
author={Wang, Qiang and He, Yuhang and Dong, Songlin and Gao, Xinyuan and Wang, Shaokun and Gong, Yihong},
booktitle={European Conference on Computer Vision},
pages={144--162},
year={2024},
organization={Springer}
}