用少量数据让SAM精准分割没见过的物体，尤其在遮挡和纹理复杂场景下表现优异。
Adapting Segment Anything Model for Unseen Object Instance Segmentation
- 通过热图生成点提示，实现无需类别信息的精准前景定位
- 仅用10%训练数据即达顶尖性能，显著降低标注成本
- 适合需要快速适配新物体的机器人视觉任务
未见物体实例分割（UOIS）对在非结构化环境中运行的自主机器人至关重要。以往方法需在大规模桌面数据集上全监督预训练才能有效。本文提出UOIS-SAM，一种数据高效的UOIS解决方案，充分利用SAM的高精度与强泛化能力。该方法集成两个核心组件：(i) 基于热图的提示生成器（HPG），生成类无关的点提示并实现精确前景预测；(ii) 层次化判别网络（HDNet），适配SAM的掩码解码器，缓解基线模型中存在的背景混淆与过度分割问题，尤其在遮挡和纹理丰富物体场景中表现突出。在OCID、OSD以及包含PhoCAL和HouseCat6D等光照挑战性数据集上的大量实验表明，即使仅使用先前方法10%的训练样本，UOIS-SAM仍能达到未见物体分割的最先进水平，验证了其在多种桌面场景下的有效性与鲁棒性。
原文摘要 · Abstract (English)
Unseen Object Instance Segmentation (UOIS) is crucial for autonomous robots operating in unstructured environments. Previous approaches require full supervision on large-scale tabletop datasets for effective pretraining. In this paper, we propose UOIS-SAM, a data-efficient solution for the UOIS task that leverages SAM's high accuracy and strong generalization capabilities. UOIS-SAM integrates two key components: (i) a Heatmap-based Prompt Generator (HPG) to generate class-agnostic point prompts with precise foreground prediction, and (ii) a Hierarchical Discrimination Network (HDNet) that adapts SAM's mask decoder, mitigating issues introduced by the SAM baseline, such as background confusion and over-segmentation, especially in scenarios involving occlusion and texture-rich objects. Extensive experimental results on OCID, OSD, and additional photometrically challenging datasets including PhoCAL and HouseCat6D, demonstrate that, even using only 10% of the training samples compared to previous methods, UOIS-SAM achieves state-of-the-art performance in unseen object segmentation, highlighting its effectiveness and robustness in various tabletop scenes.
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。