DGC-Net: Dense Geometric Correspondence Network

This is a PyTorch implementation of our work "DGC-Net: Dense Geometric Correspondence Network"

TL;DR A CNN-based approach to obtain dense pixel correspondences between two views.

Installation

• create and activate conda environment with Python 3.x

conda create -n my_fancy_env python=3.7
source activate my_fancy_env

• install Pytorch v1.0.0 and torchvision library

pip install torch torchvision

• install all dependencies by running the following command:

pip install -r requirements.txt

Getting started

• eval.py demonstrates the results on the HPatches dataset To be able to run eval.py script:

  • Download an archive with pre-trained models click and extract it to the project folder
  • Download HPatches dataset (Full image sequences). The dataset is available here at the end of the page
  • Run the following command:

  python eval.py --image-data-path /path/to/hpatches-geometry
  

• train.py is a script to train DGC-Net/DGCM-Net model from scratch. To run this script, please follow the next procedure:

  • Download the TokyoTimeMachine dataset
  • Run the command:

  python train.py --image-data-path /path/to/TokyoTimeMachine
  

Performance on HPatches dataset

Note: There is a difference in numbers presented in the original paper and obtained by the models of this repo. It might be related to the fact that both models (DGC-Net and DGCM-Net) have been trained using Pytorch v0.3.

More qualitative results are presented on the project page

License

Our code is released under the Creative Commons BY-NC-SA 3.0, available only for non-commercial use.

How to cite

If you use this software in your own research, please cite our publication:

@inproceedings{Melekhov+Tiulpin+Sattler+Pollefeys+Rahtu+Kannala:2018,
      title = {{DGC-Net}: Dense geometric correspondence network},
      author = {Melekhov, Iaroslav and Tiulpin, Aleksei and 
               Sattler, Torsten, and 
               Pollefeys, Marc and 
               Rahtu, Esa and Kannala, Juho},
       year = {2019},
       booktitle = {Proceedings of the IEEE Winter Conference on 
                    Applications of Computer Vision (WACV)}
}