The Segment Anything Model (SAM) stands as a foundational framework for image segmentation. While it exhibits remarkable zero-shot generalization in typical scenarios, its advantage diminishes when applied to specialized domains like medical imagery and remote sensing. To address this limitation, this paper introduces Conv-LoRA, a simple yet effective parameter-efficient fine-tuning approach. By integrating ultra-lightweight convolutional parameters into Low-Rank Adaptation (LoRA), Conv-LoRA can inject image-related inductive biases into the plain ViT encoder, further reinforcing SAM's local prior assumption. Notably, Conv-LoRA not only preserves SAM's extensive segmentation knowledge but also revives its capacity of learning high-level image semantics, which is constrained by SAM's foreground-background segmentation pretraining. Comprehensive experimentation across diverse benchmarks spanning multiple domains underscores Conv-LoRA's superiority in adapting SAM to real-world semantic segmentation tasks.

翻译：分割一切模型（SAM）作为图像分割的基础框架，在典型场景中展现出卓越的零样本泛化能力，但在医学影像和遥感等专业领域的应用中优势有所减弱。为解决这一局限，本文提出Conv-LoRA——一种简洁而高效的参数高效微调方法。通过将超轻量卷积参数融入低秩自适应（LoRA）框架，Conv-LoRA能够为纯视觉Transformer编码器注入图像相关的归纳偏置，进一步强化SAM的局部先验假设。值得注意的是，Conv-LoRA不仅保留了SAM丰富的分割知识，还恢复了其学习高层图像语义的能力（这一能力受到SAM前-背景分割预训练的制约）。跨多个领域的综合基准实验充分证明了Conv-LoRA在将SAM适配至实际语义分割任务中的优越性。