arXiv：cs.LG· Patricia L. Suarez, Leo Thomas Ramos, Angel D. Sappa·· 5 小时前AI 评分34
Bi-CamoDiffusion：面向伪装目标检测的边界感知扩散方法
Bi-CamoDiffusion: A Boundary-informed Diffusion Approach for Camouflaged Object Detection
AI 导读
Bi-CamoDiffusion 提出一种边界感知的扩散框架，通过无参数注入方式将边缘先验融入早期嵌入，提升边界清晰度并避免结构歧义，同时提出统一空间精度、结构约束与不确定性监督的优化目标。在 CAMO、COD10K 和 NC4K 数据集上，该模型在 $S_m$、$F_{\beta}^{w}$、$E_m$、$MAE$ 全部评测指标上超越现有 SOTA 方法，对细长结构与凸起的分割更锐利，误检更少。
正文
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, enhancing boundary sharpness and preventing structural ambiguity. An optimization objective that unifies spatial accuracy, structural constraints, and uncertainty supervision is also proposed, allowing the model to capture of both the object's global context and its intricate boundary transitions. Evaluations across the CAMO, COD10K, and NC4K datasets show that Bi-CamoDiffusion surpasses the baseline, delivering sharper delineation of thin structures and protrusions while also minimizing false positives. The model consistently outperforms existing state-of-the-art methods across all evaluated metrics, including $S_m$, $F_{\beta}^{w}$, $E_m$, and $MAE$, demonstrating a more precise object-background separation and sharper boundary recovery. Code available at: this https URL
| Comments: | 10 pages, 8 tables, 4 figures |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2603.13357 [cs.CV] |
| (or arXiv:2603.13357v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2603.13357 arXiv-issued DOI via DataCite |
|
| Journal reference: | In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1401-1410, 2026 |
Submission history
From: Leo Thomas Ramos [view email]
[v1]
Mon, 9 Mar 2026 04:01:58 UTC (6,027 KB)
[v2]
Tue, 6 Oct 2026 20:58:24 UTC (6,629 KB)
来源：arXiv：cs.LG · arxiv.org