Skip to content

Latest commit

 

History

History
77 lines (45 loc) · 1.47 KB

README.md

File metadata and controls

77 lines (45 loc) · 1.47 KB
Raw

Introduction

This is the implementation of the paper Learn to Search a Lightweight Architecture for Target-aware Infrared and Visible Image Fusion.

Requirements

  • python >= 3.6

  • pytorch == 1.7

  • torchvision == 0.8

Datasets

You can download the datasets here.

Test

python test.py

Train from scratch

step 1

python train_search.py

step 2

Find the string which descripting the searched architectures in the log file. Copy and paste it into the genotypes.py, the format should consist with the primary architecture string.

step 3

python train.py

Notes

  1. The latency lookup table is generated by:
python generate_lat_table.py
  1. The saliency map used in the loss function is obtained in advance by running:
python cal_vsm.py

Citation

If you use any part of this code in your research, please cite our paper:


@ARTICLE{9789723,  
author={Liu, Jinyuan and Wu, Yuhui and Wu, Guanyao and Liu, Risheng and Fan, Xin},  
journal={IEEE Signal Processing Letters},   
title={Learn to Search a Lightweight Architecture for Target-aware Infrared and Visible Image Fusion},   
year={2022},  
volume={},  
number={},  
pages={1-5},  
doi={10.1109/LSP.2022.3180672}}