CoCoNet

Implementation of our work:

Jinyuan Liu*, Runjia Lin*, Guanyao Wu, Risheng Liu, Zhongxuan Luo, and Xin Fan^📭, "CoCoNet: Coupled Contrastive Learning Network with Multi-level Feature Ensemble for Multi-modality Image Fusion", International Journal of Computer Vision (IJCV), 2024.

[Paper] [Arxiv]

Introduction

• Check out our recent related works 🆕:

  • 🔥 ICCV'23 Oral: Multi-interactive Feature Learning and a Full-time Multi-modality Benchmark for Image Fusion and Segmentation [paper] [code]

  • 🔥 CVPR'22 Oral: Target-aware Dual Adversarial Learning and a Multi-scenario Multi-Modality Benchmark to Fuse Infrared and Visible for Object Detection [paper] [code]

  • 🔥 IJCAI'23: Bi-level Dynamic Learning for Jointly Multi-modality Image Fusion and Beyond [paper] [code]

Installation

Clone repo:

git clone https://github.com/runjia0124/CoCoNet.git
cd CoCoNet

The code is tested with Python == 3.8, PyTorch == 1.9.0 and CUDA == 11.1 on NVIDIA GeForce RTX 2080, you may use a different version according to your GPU.

conda create -n coconet python=3.8
conda activate coconet
pip install -r requirements.txt

Quick Test

bash ./scripts/test.sh

or

python main.py \
--test --use_gpu \    
--test_vis ./TNO/VIS \
--test_ir ./TNO/IR 

To work with your own test set, make sure to use the same file names for each infrared-visible image pair if you prefer not to edit the code.

Training

Data

Get training data from [Google Drive]

Launch visdom

python -m visdom.server

Main stage training

python main.py --train --c1 0.5 --c2 0.75 --epoch 30 --bs 30 \
               --logdir <checkpoint_path> --use_gpu

Finetuning with contrastive loss

python main.py --finetune --c1 0.5 --c2 0.75 --epoch 2 --bs 30 \
               --logdir <checkpoint_path> --use_gpu

Results

Visual inspection

Down-stream task

Contact

If you have any questions about the code, please email us or open an issue,

Runjia Lin(linrunja@gmail.com) or Jinyuan Liu (atlantis918@hotmail.com).

Citation

If you find this paper/code helpful, please consider citing us:

@article{liu2023coconet,
  title={Coconet: Coupled contrastive learning network with multi-level feature ensemble for multi-modality image fusion},
  author={Liu, Jinyuan and Lin, Runjia and Wu, Guanyao and Liu, Risheng and Luo, Zhongxuan and Fan, Xin},
  journal={International Journal of Computer Vision},
  pages={1--28},
  year={2023},
  publisher={Springer}
}