MRFS_CVPR2024

The code of MRFS: Mutually Reinforcing Image Fusion and Segmentation.

@inproceedings{zhang2024mrfs,
  title={MRFS: Mutually Reinforcing Image Fusion and Segmentation},
  author={Zhang, Hao and Zuo, Xuhui and Jiang, Jie and Guo, Chunchao and Ma, Jiayi},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={26974--26983},
  year={2024}
}

#### Recommended Environment:
- [ ] python = 3.8 - [ ] torch = 1.8.1+cu111 - [ ] timm = 0.9.8 - [ ] numpy = 1.24.4 - [ ] scipy = 1.10.1 - [ ] pillow = 10.1.0 - [ ] tqdm = 4.66.1 - [ ] tensorboardX = 2.6.2.2 - [ ] opencv-python = 4.8.1.78

Training:

• Prepare training data & set the training parameters:
  
  • Dataset fomula:
  • Dataset name
    ------RGB folder
    ------THE folder
    ------Label folder
    ------train.txt
    ------val.txt
    ------test.txt
    
• Run CUDA_VISIBLE_DEVICES="GPU IDs" python -m torch.distributed.launch --nproc_per_node="GPU numbers you want to use" train.py
  
• Example CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2 train.py
  

Test:

• pretrained weights can be found at here: FMB, MFNet, PST900, place the checkpoints under the corrspanding floder.
• Set the testing parameters:
  
• Run CUDA_VISIBLE_DEVICES="GPU IDs" python eval.py -e="epoch"
  
• Example CUDA_VISIBLE_DEVICES=0 python eval.py -e=MRFS