SD-IPC

---

SD-IPC (NeurIPS 2023)

The CLIP Model is Secretly an Image-to-Prompt Converter

Yuxuan Ding¹, Chunna Tian¹, Haoxuan Ding², Lingqiao Liu³

¹ Xidian University, ² Northwestern Polytechnical University, ³ The University of Adelaide

News

• News (2024-03-09): We have released the code for SD-IPC.
• News (2023-09-22): 🎉🎉🎉 Congratulations on SD-IPC being accepted by NeurIPS 2023!

Todo

• release the the camera-ready version of the paper and supplementary materials
• release the code for fine-tuning SD-IPC (SD-IPC-FT, SD-IPC-CT)

Acknowledgments

Our work is standing on the shoulders of giants. We want to thank the following contributors that our code is based on:

• open-source library of diffusion models:
  • (🤗 Diffusers) https://github.com/huggingface/diffusers
  • (Stable Diffusion) https://github.com/CompVis/stable-diffusion
• open-source library of transformer models:
  • (🤗 Transformers) https://github.com/huggingface/transformers

Citation

If this work is helpful for your research, please consider citing the following BibTeX entry.

@article{ding2024clip,
  title={The CLIP Model is Secretly an Image-to-Prompt Converter},
  author={Ding, Yuxuan and Tian, Chunna and Ding, Haoxuan and Liu, Lingqiao},
  journal={Advances in Neural Information Processing Systems},
  volume={36},
  year={2024}
}