Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Bi-CamoDiffusion: A Boundary-informed Diffusion Approach for Camouflaged Object Detection

Patricia L. Suarez, Leo Thomas Ramos, Angel D. Sappa

Abstract

Bi-CamoDiffusion is introduced, an evolution of the CamoDiffusion framework for camouflaged object detection. It integrates edge priors into early-stage embeddings via a parameter-free injection process, which enhances boundary sharpness and prevents structural ambiguity. This is governed by a unified optimization objective that balances spatial accuracy, structural constraints, and uncertainty supervision, allowing the model to capture of both the object's global context and its intricate boundary transitions. Evaluations across the CAMO, COD10K, and NC4K benchmarks show that Bi-CamoDiffusion surpasses the baseline, delivering sharper delineation of thin structures and protrusions while also minimizing false positives. Also, our model consistently outperforms existing state-of-the-art methods across all evaluated metrics, including $S_m$, $F_β^{w}$, $E_m$, and $MAE$, demonstrating a more precise object-background separation and sharper boundary recovery.

Bi-CamoDiffusion: A Boundary-informed Diffusion Approach for Camouflaged Object Detection

Abstract

Bi-CamoDiffusion is introduced, an evolution of the CamoDiffusion framework for camouflaged object detection. It integrates edge priors into early-stage embeddings via a parameter-free injection process, which enhances boundary sharpness and prevents structural ambiguity. This is governed by a unified optimization objective that balances spatial accuracy, structural constraints, and uncertainty supervision, allowing the model to capture of both the object's global context and its intricate boundary transitions. Evaluations across the CAMO, COD10K, and NC4K benchmarks show that Bi-CamoDiffusion surpasses the baseline, delivering sharper delineation of thin structures and protrusions while also minimizing false positives. Also, our model consistently outperforms existing state-of-the-art methods across all evaluated metrics, including , , , and , demonstrating a more precise object-background separation and sharper boundary recovery.
Paper Structure (24 sections, 19 equations, 4 figures, 9 tables)

This paper contains 24 sections, 19 equations, 4 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Proposed approach
  4. Overview
  5. Edge prior construction and feature injection
  6. Grayscale conversion
  7. Sobel edge magnitude
  8. Parameter-free feature modulation
  9. Boundary-sharpening pre-filter
  10. Multi-scale RGB-Edge Consistency Loss
  11. Focal-structure loss
  12. Ground-truth edge loss
  13. Uncertainty-aware loss
  14. RGB-Edge Consistency Loss
  15. Multi-scale aggregation and total objective
  16. ...and 9 more sections

Figures (4)

  • Figure 1: High-level overview of the proposed Bi-CamoDiffusion. Components highlighted in orange represent the contributions introduced in this work, while purple denotes the original CamoDiffusion components.
  • Figure 2: Qualitative comparison between our boundary-informed method and the baseline on the CAMO dataset.
  • Figure 3: Qualitative comparison between our boundary-informed method and the baseline on the COD10K dataset.
  • Figure 4: Qualitative comparison between our boundary-informed method and the baseline on the NC4K dataset.