Learning Conditional Invariance through Cycle Consistency

This repository provides a basic TensorFlow 1 implementation of the proposed model in our GCPR 2021 paper "Learning Conditional Invariance through Cycle Consistency".

Check out our talk given at GCPR 2021 (on Youtube) for an overview of the paper!

Executing the Code

The developed code used the following dependencies:

  - python=3.6.12
  - matplotlib=3.3.2
  - tensorflow-gpu=1.14

You can install a corresponding environment with

conda env create -f requirements.yml

activate the environment

conda activate CondInvCC

and execute the script with

python main.py --mode train --experiment ellipse

for training our model in the ellipse setting.

Pre-trained Models

We provide the pretrained models for the ellipse and ellipsoid experiment which you can execute with

python main.py --mode test --experiment ellipsoid

Reference

If you like our paper and use it for your research, please cite us.

@inproceedings{SamarinNesterov2021,
  title={Learning Conditional Invariance through Cycle Consistency},
  author={Samarin*, Maxim and Nesterov*, Vitali and Wieser, Mario and Wieczorek, Aleksander
   and Parbhoo, Sonali and Roth, Volker},
  booktitle={German Conference on Pattern Recognition},
  year={2021},
  organization={Springer}
}