princetonvisualai / conceptmix

10 stars 2 forks source link

ConceptMix

ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty

NeurIPS 2024, Dataset and Benchmark Track [paper][website]

Xindi Wu*, Dingli Yu*, Yangsibo Huang*, Olga Russakovsky, Sanjeev Arora

We propose ConceptMix, a scalable and customizable benchmark for evaluating the compositional capabilities of T2I models, including prompts from 8 visual concept categories.

News 🔥

Table of Contents

Installation

To set up the environment for ConceptMix, you can use the provided environment.yml file to create a Conda environment.

conda env create -f environment.yml
conda activate conceptmix

Usage

Step 1: Generating Prompts

To generate text, use the generate_text.sh script.

sh ./generate_text.sh

Step 2: Generating Images

To generate images, use the generate_image.sh script.

sh ./generate_image.sh

Grading

To evaluate the generated images, use the grading.sh script.

sh ./grading.sh

You can also submit the grading job using the grading.slurm script if you are using a SLURM-based cluster.

sbatch grading.slurm

Visual Concepts

The configuration files are located in the config directory. You can modify these files to adjust the settings for prompt generation.

Contributing

We welcome contributions from the community. If you would like to contribute, please fork the repository and submit a pull request.

License

This project is licensed under the MIT License.

Citation

If you find this repository useful for your research, please cite with the following BibTeX entry:

@article{wu2024conceptmix,
  title={ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty},
  author={Wu, Xindi and Yu, Dingli and Huang, Yangsibo and Russakovsky, Olga and Arora, Sanjeev},
  journal={arXiv preprint arXiv:2408.14339},
  year={2024}
}

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No.2107048. 
Any opinions, findings, and conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. 
DY and SA are supported by NSF and ONR. YH is supported by the Wallace Memorial Fellowship. 
We thank many people for their helpful discussion, feedback, and human studies listed in alphabetical order by last name: 
Allison Chen, Jihoon Chung, Victor Chu, Derek Geng, Luxi He, Erich Liang, Kaiqu Liang, Michel Liao, Yuhan Liu, Abhishek Panigrahi, Simon Park, Ofir Press, Zeyu Wang, Boyi Wei, David Yan, William Yang, Zoe Zager, Cindy Zhang, and Tyler Zhu from Princeton University, 
Zhiqiu Lin, Tiffany Ling from Carnegie Mellon University, 
Chiyuan Zhang from Google Research.