双层适配器提升无提示曲线结构分割
Dual-level Adapter Boosting Prompt-free Curvilinear Structure Segmentation
在冻结的 SAM 上插入块内与块间双层适配器，实现无提示、仅需 18 张标注图像的跨域曲线结构分割，可直接迁移到医学/遥感细长目标（血管、道路、纤维）的少样本分割场景。
原文依据 SACM introduces a dual-level adapter architecture that enables both fine-grained and domain-adaptive enhancement
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论曲线结构分割在医学影像、遥感与材料科学等领域至关重要。现有方法往往需要大量领域特定的训练，且难以泛化到新领域。为克服这些局限，我们提出 Segment Anything Curve Model（SACM）——一个构建于预训练 Segment Anything Model（SAM）之上的通用曲线分割框架。SACM 引入双层适配器架构，可同时实现细粒度增强与领域自适应增强：块级内部适配器细化局部结构表征，外部适配器则促进跨域特征对齐。具体而言，内部适配器嵌入每个 Transformer 块内部，针对细薄而复杂的曲线模式对特征进行局部适配与细化；外部适配器则跨块运作，以捕获全局的多层上下文信息并促进领域自适应。此外，SACM 引入一种特征融合机制，聚合来自所有外部适配器的多层特征，并通过前馈网络（FFN）模块进行融合；同时在掩码解码器中引入双阶段细化过程，以增强拓扑结构与连通性。该设计实现了无提示、数据高效的微调，并且仅用 18 张标注图像训练即可获得稳健的跨域泛化能力。在十二个多样化曲线数据集上的大量实验验证，SACM 达到了最优性能。
英文原文摘要 1515 字符
Curvilinear structure segmentation is essential in domains such as medical imaging, remote sensing, and materials science. Existing methods often require extensive domain-specific training and lack generalization to novel domains. To overcome these limitations, we propose the Segment Anything Curve Model (SACM) -- a universal, curvilinear segmentation framework built upon the pretrained Segment Anything Model (SAM). SACM introduces a dual-level adapter architecture that enables both fine-grained and domain-adaptive enhancement: block-level internal adapters refine local structural representations, while external adapters facilitate cross-domain feature alignment. Specifically, the internal adapters are embedded within each Transformer block to locally adapt and refine features for thin and intricate curvilinear patterns, while the external adapters operate across blocks to capture global, multi-layer contextual information and facilitate domain adaptation. Furthermore, SACM introduces a feature fusion mechanism that aggregates multi-layer features from all external adapters and fuses them via a Feed-Forward Network (FFN) module, and a dual-stage refinement process in the mask decoder to enhance topology and connectivity. This design enables prompt-free, data-efficient fine-tuning and achieves robust cross-domain generalization when trained with only 18 annotated images. Extensive experiments across twelve diverse curvilinear datasets validate that SACM achieves state-of-the-art performance.