Home

Awesome

RAVEN-FAIR

Balanced RAVEN dataset from the paper: 'Scale-Localized Abstract Reasoning', presented at CVPR 2021.

Paper Code

Requirements

Tested on both linux and windows 10.

Generating the dataset

To create the dataset, run:

python main.py --fair FAIR --save-dir DEST

Original RAVEN will be created at <DEST>/RAVEN. RAVEN-FAIR will be created at <DEST>/RAVEN-F.

Acknowledgement

We thank the original creators of the RAVEN dataset: Chi Zhang, Feng Gao, Baoxiong Jia, Yixin Zhu, Song-Chun Zhu. The original code can be found at the repository: RAVEN.

Citation

We thank you for showing interest in our work. If our work was beneficial for you, please consider citing us using:

@inproceedings{benny2021scale,
  title={Scale-localized abstract reasoning},
  author={Benny, Yaniv and Pekar, Niv and Wolf, Lior},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={12557--12565},
  year={2021}
}

If you have any question, please feel free to contact us.