通过深度图先验提升U-Net在肠镜图像中息肉分割精度
Learning with Geometric Priors in U-Net Variants for Polyp Segmentation
- 用模拟肠镜深度数据微调VGGT，生成专属息肉深度图
- 将深度图作为几何先验注入编码器特征，提升低对比度场景表现
- 模块即插即用，适用于多种U-Net变体，适合医学图像分割研究者
准确可靠的息肉分割对早期结直肠癌检测和计算机辅助诊断至关重要。尽管基于卷积神经网络、Transformer和Mamba的U-Net变体已取得良好性能，但在低对比度或杂乱的肠镜图像中仍难以捕捉几何与结构线索。为此，我们提出一种新型几何先验引导模块（GPM），将显式几何先验注入基于U-Net的架构中。具体地，我们在模拟的ColonDepth数据集上微调视觉几何基础Transformer（VGGT），以生成适配内镜域的息肉图像深度图。这些深度图经由GPM处理后，被编码为编码器特征图中的几何先验，并通过空间与通道注意力机制进一步优化，突出局部空间与全局通道信息。GPM可即插即用，无缝集成于多种U-Net变体。在五个公开息肉分割数据集上的大量实验表明，其性能持续优于三个强基线模型。代码与生成的深度图已开源。
原文摘要 · Abstract (English)
Accurate and robust polyp segmentation is essential for early colorectal cancer detection and for computer-aided diagnosis. While convolutional neural network-, Transformer-, and Mamba-based U-Net variants have achieved strong performance, they still struggle to capture geometric and structural cues, especially in low-contrast or cluttered colonoscopy scenes. To address this challenge, we propose a novel Geometric Prior-guided Module (GPM) that injects explicit geometric priors into U-Net-based architectures for polyp segmentation. Specifically, we fine-tune the Visual Geometry Grounded Transformer (VGGT) on a simulated ColonDepth dataset to estimate depth maps of polyp images tailored to the endoscopic domain. These depth maps are then processed by GPM to encode geometric priors into the encoder's feature maps, where they are further refined using spatial and channel attention mechanisms that emphasize both local spatial and global channel information. GPM is plug-and-play and can be seamlessly integrated into diverse U-Net variants. Extensive experiments on five public polyp segmentation datasets demonstrate consistent gains over three strong baselines. Code and the generated depth maps are available at: https://github.com/fvazqu/GPM-PolypSeg
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。