Skip to content
/ SFDFusion Public

Code of "SFDFusion: An Efficient Spatial-Frequency Domain Fusion Network For Infrared and Visible Image Fusion"

Notifications You must be signed in to change notification settings

lqz2/SFDFusion

Repository files navigation

SFDFusion

Code of "SFDFusion: An Efficient Spatial-Frequency Domain Fusion Network for Infrared and Visible Image Fusion"

Accepted by ECAI 2024

paper link [here]

Environments

python 3.10
cuda 11.8

Install

conda create -n SFD python=3.10
conda activate SFD
pip install -r requirements.txt

Train

The training process needs wandb API key. The config file is ./configs/cfg.yaml

python train.py

Inference

python fuse.py

Dataset

MSRS dataset is used to train. You can get it from here.

About

Code of "SFDFusion: An Efficient Spatial-Frequency Domain Fusion Network For Infrared and Visible Image Fusion"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages