Boosting Foreground-Background Disentanglement for Camouflaged Object Detection

February 21, 2026 · View on GitHub

ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 2025

Authors:
Jiesheng Wu¹ (Corresponding author), Fangwei Hao², Jing Xu²
¹ School of Computer and Information, Anhui Normal University, Wuhu, China ² College of Artificial Intelligence, Nankai University, Tianjin, China

1. Preface

  • This repository provides code for "Boosting Foreground-Background Disentanglement for Camouflaged Object Detection" ACM TOMM 2025. Paper

2. Overview

2.1. Introduction

In nature, certain objects exhibit patterns that closely resemble their backgrounds, a phenomenon commonly referred to as Camouflaged Object Detection (COD). We argue that existing COD approaches often suffer from insufficient discriminability for these objects, which we attribute to a lack of effective disentangling of foreground and background representations. To address this, we propose a novel Foreground-Background Disentanglement Network (FBD-Net) that enhances foreground-background disentanglement learning to improve discriminability. Specifically, we design an Edge-guided Foreground-Background Decoupling (EFBD) module, which facilitates the separated learning of foreground and background representations. Additionally, we introduce the Foreground-Background Representation Disentangling Head (DisHead) to further boost the discriminative power of the model. The DisHead consists of two objectives: the Edge Objective and the FoBa Objective. Furthermore, we propose three complementary modules: the Context Aggregation Module (CAM) for initial coarse object detection, the Scale-Interaction Enhanced Pyramid (SIEP) for multi-scale information extraction, and the Cross-Stage Adaptive Fusion (CSAF) module for subtle clue accumulation. Extensive experiments demonstrate that both our CNN-based and Transformer-based FBD-Nets outperform 26 state-of-the-art COD methods across four public datasets.

2.2. Framework Overview


Figure 1: Overall architecture of the proposed FBD-Net. FBD-Net consists of five key components: a CAM, a SIEP, an EFBD module, a CSAF module, and a DisHead.

2.3. Quantitative Results


Figure 2: Quantitative Results

2.4. Qualitative Results


Figure 3: Qualitative Results.

3. Proposed Method

3.1. Training/Testing

The training and testing experiments are conducted using PyTorch with two NVIDIA Tesla V100 GPUs of 32 GB Memory.

  1. Configuring your environment (Prerequisites):

    • Installing necessary packages: pip install -r requirements.txt.

  2. Downloading necessary data:

    • downloading training dataset and move it into ./data/, which can be found from Baidu Drive (extraction code: ekd2).

    • downloading testing dataset and move it into ./data/, which can be found from Baidu Drive (extraction code: nhwe).

    • downloading our weights and move it into ./save_models/PVT-V2-B4-384.pth, which can be found from (Baidu Drive) (extraction code: 2855).

    • downloading weights and move it into ./pre_train/pvt_v2_b4.pth, which can be found from Baidu Drive (extraction code: u1u6).

  3. Training Configuration:

    • After you download training dataset, just run MyTrain.py to train our model.

  4. Testing Configuration:

    • After you download all the pre-trained model and testing dataset, just run MyTest.py to generate the final prediction maps.

    • You can also download prediction maps and edge prediction maps ('CHAMELEON', 'CAMO', 'COD10K', 'NC4K') from Baidu Drive (extraction code: w2mw)).

Note: If you have difficulty accessing Baidu Drive, please contact us for alternative download links.

3.2 Evaluating your trained model:

One evaluation is written in Python codes (link), or Matlab codes (link). please follow this the instructions in MyEval.py and just run it to generate the evaluation results.

4. Citation

Please cite our paper if you find the work useful, thanks!

@article{wu2025boosting,
  title={Boosting Foreground-Background Disentanglement for Camouflaged Object Detection},
  author={Wu, Jiesheng and Hao, Fangwei and Xu, Jing},
  journal={ACM Transactions on Multimedia Computing, Communications and Applications},
  volume={21},
  number={12},
  pages={1--23},
  year={2025},
  publisher={ACM New York, NY}
}

5. Contact

For any questions, discussions, or collaboration opportunities, please contact:

Jiesheng Wu
School of Computer and Information, Anhui Normal University, Wuhu, China Email: jasonwu@mail.nankai.edu.cn

⬆ back to top