/MaxCon_RL

This is demo code of our paper in CVPR 2021

Primary LanguagePythonMIT LicenseMIT

Demo -- Unsupervised Learning for Robust Fitting

This is demo of the following CVPR 2021 paper: Unsupervised Learning for Robust Fitting: A Reinforcement Learning Approach

Paper can be accessed at: https://arxiv.org/abs/2103.03501

Pre-requisite

- Pytorch
- Torch geometric
- cvxpy: pip install cvxpy

Usage

  • Run the inference using pre-trained model:
python inference.py
  • Download minimal data and train the model
python downloadData.py
python train.py

Citation

If you find our work useful in your research, please consider citing:

@inproceedings{MaxCon_RL,
  title={Unsupervised Learning for Robust Fitting: A ReinforcementLearning Approach},
  author={Giang Truong and Huu Le and David Suter and Erchuan Zhang and Syed Zulqarnain Gilani},
  booktitle={CVPR 2021},
  year={2021}
}