PSCM in Open-Stylized Image Captioning

Code release for the paper "Plugging Stylized Controls in Open-Stylized Image Captioning"

Changelog

• 2025/05/29 upload the code.

Requirements

• python>=3.8
• PyTorch 1.2.0
• torchvision
• scikit-image=0.18.1

Data

• Download datasets
• Extract them to data/SentiCap/, data/FlickrStyle10K/ and data/News/, respectively.
• Split the dataset into train and test folder

Training

• python train_PSCM.py

Citation

If you find this paper useful in your research, please consider citing:

@inproceedings{wang2023plugging,
  title={Plugging Stylized Controls in Open-Stylized Image Captioning},
  author={Wang, Jie and Zheng, Yixiao and Du, Ruoyi and Zhang, Yiming and Liang, Kongming and Ma, Zhanyu},
  booktitle={Chinese Conference on Pattern Recognition and Computer Vision (PRCV)},
  pages={309--320},
  year={2023},
  organization={Springer}
}

Contact

Thanks for your attention! If you have any suggestion or question, you can leave a message here or contact us directly:

• wangjiee@bupt.edu.cn
• mazhanyu@bupt.edu.cn