🔥 [ACM MM 2025] From Language to Instance: Generative Visual Prompting for Zero-shot Camouflaged Object Detection
Code release of paper:
From Language to Instance: Generative Visual Prompting for Zero-shot Camouflaged Object Detection
- [2025.10.29] The code of Lip is released!
- [2025.07.05] Lip is accepted to ACM MM 2025!
- Download the datasets from the follow links:
Camouflaged Object Detection Dataset
- Put it in data/TestDataset
- My code was implemented with Python 3.10 and PyTorch 2.1.0. We recommend creating environment and installing all the dependencies, as follows:
# create virtual environment
conda create --name Lip python=3.10
conda activate Lip
# prepare codebase
git clone https://github.com/skywalker0523/Lip.git
cd Lip
# prepare environment
pip install -r requirement.txt
# prepare dinov2
git clone https://github.com/facebookresearch/dinov2.git
# prepare SAM
pip install git+https://github.com/facebookresearch/segment-anything.git
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth- Our Lip is a training-free test-time adaptation approach, so you can play with it by running:
python main.py --config config/CHAMELEON.yaml If you find our work useful in your research, please consider citing:
@inproceedings{zhang2025language,
title={From Language to Instance: Generative Visual Prompting for Zero-shot Camouflaged Object Detection},
author={Zhang, Zihou and Li, Hao and Yang, Zhengwei and Hu, Zechao and Li, Liang and Wang, Zheng},
booktitle={Proceedings of the 33rd ACM International Conference on Multimedia},
pages={382--391},
year={2025}
}