Exploring Deeper! Segment Anything Model with Depth Perception for Camouflaged Object Detection

Zhenni Yu, Xiaoqin Zhang, Li Zhao, Yi Bin, Guobao Xiao ACM MM, 2024

Usage

Installation

git clone https://github.com/guobaoxiao/DSAM
cd DSAM

environment

conda env create -f environment.yaml

From datasets to npz

you can load down the COD datasets and run this to get npz for train.

python pre_npz.py

• COD datasets: download the COD datasets set from here(CAMO, COD10K, NC4K), and put into 'data/'

• depth datasets: download the depth datasets set, put into 'data/'. The depth image is from PopNet.

  • 通过百度网盘分享的文件：Train_depth.zip 链接：https://pan.baidu.com/s/1grcASolza9GLpHIVk8mESQ 提取码：wocz
  • 通过百度网盘分享的文件：Test_depth.zip 链接：https://pan.baidu.com/s/1HobAvMBpfSUfUHNXGZeFLw 提取码：32ut

Weights

• pre-weigth: download the weight of sam from here, the weight of pvt form xxx, put into 'work_dir_cod/SAM/'

• DSAM: download the weight of well-trained DSAM, put into 'work_dir_cod/DSAM'

  • 通过百度网盘分享的文件：DSAM.pth 链接：https://pan.baidu.com/s/1148mXSjTv7OKlWHcfZFh5A 提取码：39xx

The predicted image

• DSAM:
  • 通过百度网盘分享的文件：DSAM.zip 链接：https://pan.baidu.com/s/1V5372Z_GdHzYEyOR3iEu4Q 提取码：fu49

Train

python Mytrain.py

Test

python Mytest.py

Translate npz to img

python transformer_nzp_2_gt.py

eval

python MSCAF_COD_evaluation/evaluation.py

Citation

If you find this project useful, please consider citing:

@inproceedings{yu2024exploring,
title={Exploring Deeper! Segment Anything Model with Depth Perception for Camouflaged Object Detection},
author={Zhenni Yu and Xiaoqin Zhang and LiZhao and Yi Bin and Guobao Xiao},
booktitle={ACM Multimedia 2024},
year={2024},
url={https://openreview.net/forum?id=d4A0Cw1gVS}
}