Skip to content

[IJCAI 2023] A Large-scale Film Style Dataset for Learning Multi-frequency Driven Film Enhancement

License

Notifications You must be signed in to change notification settings

CXH-Research/FilmNet

Repository files navigation

Zinuo Li 👨‍💻‍ , Xuhang Chen 👨‍💻‍ , Shuqiang Wang 📮 and Chi-Man Pun 📮 ( 👨‍💻‍ Equal contributions, 📮 Corresponding )

University of Macau, SIAT CAS

2023 International Joint Conference on Artificial Intelligence (IJCAI 2023)

image

🔮 Important news

[11/03/2023:] There was a typo regarding data for the Cinema-SSIM of DeepLPF, which we have corrected in the arxiv version of the paper.

⚙️ Usage

Installation

git clone https://github.com/CXH-Research/FilmNet.git
cd FilmNet
pip install -r requirements.txt

Training

Please first specify TRAIN_DIR, VAL_DIR and SAVE_DIR in section TRAINING in traning.yml

For single GPU training:

python train.py

For multiple GPUs training:

accelerate config
accelerate launch train.py

If you have difficulties on the usage of accelerate, please refer to Accelerate.

Inference

Please first specify TRAIN_DIR, VAL_DIR and SAVE_DIR in section TESTING in traning.yml

python test.py

🛎 Citation

If you find our work helpful for your research, please cite:

@inproceedings{ijcai2023p129,
  title     = {A Large-Scale Film Style Dataset for Learning Multi-frequency Driven Film Enhancement},
  author    = {Li, Zinuo and Chen, Xuhang and Wang, Shuqiang and Pun, Chi-Man},
  booktitle = {Proceedings of the Thirty-Second International Joint Conference on
               Artificial Intelligence, {IJCAI-23}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Edith Elkind},
  pages     = {1160--1168},
  year      = {2023},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2023/129},
  url       = {https://doi.org/10.24963/ijcai.2023/129},
}