无需标注数据，一键分割植物3D器官。
Zero-Shot 3D Plant Organ Segmentation with SAM3 and Semantic NeRFs
- 用文本提示SAM3与语义NeRF融合，实现零样本分割
- 在10种植物上平均达0.856 mIoU，叶片分割超0.91
- 适合无标注数据的植物表型分析研究者
精准的3D植物器官分割是自动化表型分析的基础。现有方法依赖标注数据或物种特异性模型配置。我们提出一种无需标注的3D植物器官分割流程，结合文本提示的SAM3分割与语义神经辐射场（Semantic NeRFs）。仅需多视角RGB图像和类别名称列表，该零样本流程即可生成无手动标注、无需物种微调或领域特定预处理的语义3D点云。多视角NeRF融合作为有效的隐式共识机制，将不完美的单帧掩码提升为精确的3D标签。在受控的Begonia maculata测试平台上，SAM3流程达到92.6% mIoU，达到使用完美真值掩码所建立的95.9%基准上限。该流程在包含十种多样化植物点云的新数据集上进一步评估，平均mIoU为0.856，每种植物的叶和盆体IoU均高于0.91和0.90。结果表明，无需标注的3D植物器官分割已可实现，并接近监督方法水平。
原文摘要 · Abstract (English)
Accurate 3D plant organ segmentation is fundamental to automated phenotyping. Existing approaches rely on annotated training data or species-specific model configurations. We present an annotation-free pipeline for 3D plant organ segmentation, combining text-prompted SAM3 segmentation with semantic neural radiance fields (NeRFs). Given only multi-view RGB images and a list of class names, our zero-shot pipeline produces semantically labeled 3D point clouds without manual annotation, per-species fine-tuning, or domain-specific preprocessing. Multi-view NeRF fusion acts as effective implicit consensus mechanism that lifts imperfect per-frame masks into accurate 3D labels. On a controlled Begonia maculata testbed the SAM3 pipeline achieves 92.6% mIoU, reaching 95.9% of the oracle upper bound established with perfect ground-truth masks. The pipeline was further evaluated on a new dataset spanning ten diverse plant point clouds reaching an average 0.856 mIoU, with leaf and pot IoU above 0.91 and 0.90 for every species, respectively. These results demonstrate that annotation-free 3D plant organ segmentation is now feasible and approaching the range of supervised methods.
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。