README.md

April 27, 2025 ยท View on GitHub

Open-source schedule

  • 28 Jun 2024: Open-sourced model and test code
  • 19 Aug 2024: Released prediction map, pretrained model, backbone, and datasets
  • 27 April 2025: Training code has been uploaded

Downloads

ItemBaidu PanGoogle Drive
Prediction MapLinkLink
Pretrained ModelLinkLink
Backbone ModelLinkLink
DatasetsLinkLink

environment

  1. Python 3.8.13
  2. others packages can be found at requirement.txt

start

  1. git clone https://github.com/CHANTILLY2023/MAMIFNet

  2. conda create --name MAMIFNet python=3.8.12

  3. conda activate MAMIFNet

  4. pip install torch==1.9.1+cu111 torchvision==0.10.1+cu111 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html

  5. pip install -r requirements.txt

  6. pip install -U openmim

  7. mim install mmcv-full==1.5.2

test

The relevant setting paths for data and weights are located in config.py. The default test is as follows:

  1. create folder './data', download datasets file COD_dataset at https://pan.baidu.com/s/1Kpxaqv0n5YP-kDT8Id4wzA?pwd=pugv, unzip and move it to './data'

  2. download backbone model at https://pan.baidu.com/s/16bKJaxewVfFXd73ICV6zDg?pwd=vn6v and unzip it.

  3. download pretrained model at https://pan.baidu.com/s/10gWYyESaV6LaCZv8_1PVBA?pwd=z8d1 and unzip it.

  4. run test.py