Skip to content
/ e2fai Public

Project page of the paper: "Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras"

Notifications You must be signed in to change notification settings

tub-rip/e2fai

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras

Official repository for Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras, ICCV 2025 highlight, by Shuang Guo, Friedhelm Hamann and Guillermo Gallego.

Also known as E2FAI: Events to Flow And Intensity

Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras

Citation

If you use this work in your research, please cite it as follows:

@InProceedings{Guo25iccv,
  author        = {Shuang Guo and Friedhelm Hamann and Guillermo Gallego},
  title         = {Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras},
  booktitle     = {IEEE/CVF International Conference on Computer Vision (ICCV)},
  year          = 2025
}

Setup

High-level Input-Output

Input:

  • Events.

Output:

  • Optical flow.
  • Intensity image.

Environment Setup

We recommend using conda to set up the environment.

Create the environment using the provided environment.yml:

conda env create -f environment.yml
conda activate e2fai

Pretrained Model

Download the pretrained model from here.

Inference on DSEC

python inference/inference_dsec.py --config config/dsec.yaml --ckp_path path_to_ckpt --name test_name --gpu 0

Related Works

Additional Resources

About

Project page of the paper: "Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages