michael861227/pinet

Finding the Achilles Heel: Progressive Identification Network for Camouflaged Object Detection, IEEE International Conference on Multimedia & Expo (ICME), 2022. (accepting rate = 29%)

(ICME 2022) Finding the Achilles Heel: Progressive Identification Network for Camouflaged Object Detection

Paper for our Progressive Identification Network, also called PINet [PDF], published at ICME 2022

Model Architecture

Prerequisites

• Install Enviroment

    conda create -n PINet python=3.7
    conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge
    pip install tensorboardX
    pip install opencv-python

• Install Apex

    git clone https://github.com/NVIDIA/apex.git
    cd apex
    python setup.py install --cpp_ext

Datasets

• Training data from Google Drive
• Testing data from Google Drive

Pretrained models

You can download our pretrained model from Google Drive

Usage

For training, use the command:

python train.py

For testing, use the command:

python test.py

Results

Citation

If you find our paper useful in your research, please cite us using the following entry:

@INPROCEEDINGS{9859854,
  author={Chou, Mu-Chun and Chen, Hung-Jen and Shuai, Hong-Han},
  booktitle={2022 IEEE International Conference on Multimedia and Expo (ICME)}, 
  title={Finding the Achilles Heel: Progressive Identification Network for Camouflaged Object Detection}, 
  year={2022},
  volume={},
  number={},
  pages={1-6},
  doi={10.1109/ICME52920.2022.9859854}}