hyzcn / TVR

Code for paper "Transformation Driven Visual Reasoning".

TVR - Transformation driven Visual Reasoning

| Project | Paper |

Task: Given both the initial and final states, the target is to infer the corresponding single-step or multi-step transformation.

Dataset Generation

Please see the README.md for data generation.

Model Training and Evaluation

Coming soon...

Citing TVR

If you find TVR useful for your research then please cite:

@article{hong2020tvr,
    title={Transformation Driven Visual Reasoning},
    author={Xin Hong, Yanyan Lan, Liang Pang, Jiafeng Guo and Xueqi Cheng},
    year = {2020},
    journal={arXiv}
}

LICENSE

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

Notice: Some materials are directly inherited from CLEVR which are licensed under BSD License. More details can be found in this document.