Home

Awesome

UEDG:Uncertainty-Edge Dual Guided Camouflage Object Detection

Title: Uncertainty-Edge Dual Guided Camouflage Object Detection
Paper link: Early access version avalible here.

1. News

[2023-07-11] Code and corresponding data is uploaded with user guide.
[2023-07-03] Paper has been accepted by IEEE Transaction on Multimedia. :partying_face: Congradulations!!! :partying_face:
[2023-05-27] Detection results on four dataset: CHAMELEON, CAMO, COD10K-test, and NC4K are avilible: Google Drive or Baidu Netdisk(9cu6).
[2023-05-26] Initial repository.
[2022-11-22] Manuscript uploaded.

2. Features

<p align="center"> <img src="assest/features.png"/> <br/> <em> Figure 1: In this paper, we present the visualization results of the edge and uncertainty guidance operation in a highly challenging scenario. The red boxes indicate regions where the indistinguishable parts yield higher uncertainty scores. By incorporating edge information, our UEDG (Uncertainty Edge Guidance) approach achieves favorable performance. </em> </p>

3. Overview

<p align="center"> <img src="assest/overview.png"/> <br/> <em> Figure 2: UEDG strucuture overview. </em> </p> <p align="center"> <img src="assest/qualitative results.png"/> <br/> <em> Figure 3: UEDG strucuture overview. </em> </p> <p align="center"> <img src="assest/quantitative results.png"/> <br/> </p>

4. Usage

4.1. Requirements

conda create -n UEDG python=3.8
conda activate UEDG
pip install -r ./requirements.txt

For cuda11.6:

conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.6 -c pytorch -c conda-forge

4.2 Data Preparation

<!-- + downloading pretrained weights `Net_epoch_best.pth` and move it into `./log/Mytrain/`, which can be found in [GoogleDrive](https://drive.google.com/file/d/1PD2mexy-IwnyWsb7WF59V1VAF24UKOcu/view?usp=sharing). -->

4.2. Training

python MyTrain.py

4.3. Testing

python MyTest.py

5. Thanks

Code copied a lot from HVision-NKU/CamoFormer, thograce/BGNet, weijun88/F3Net, HUuxiaobin/HitNet, GewelsJI/DGNet, clelouch/BgNet, fanyang587/UGTR, whai362/PVT. Thanks for their great works!

6. License

The source code is free for research and education use only. Any commercial usage should get formal permission first.

7.Citation

@ARTICLE{10183371,
author={Lyu, Yixuan and Zhang, Hong and Li, Yan and Liu, Hanyang and Yang, Yifan and Yuan, Ding},
journal={IEEE Transactions on Multimedia}, 
title={UEDG:Uncertainty-Edge Dual Guided Camouflage Object Detection}, 
year={2023},
doi={10.1109/TMM.2023.3295095}}