[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2783":3},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":23,"tags":25,"view_count":31,"doi":32,"paper":33,"created_at":67},2783,"Toward scalable organ-level 3D plant segmentation: A systematic and quantitative review through the lens of the data-algorithm-computing triangle","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.isprsjprs.2026.09.003","The precise characterization of plant morphology provides valuable insights into plant-environment interactions and genetic evolution. A key technology for extracting this information is 3D segmentation, which delineates individual plant organs from complex point clouds. Despite significant progress in general 3D computer vision domains, the adoption of 3D segmentation for plant phenotyping remains limited by three major challenges: (i) the scarcity of large-scale annotated datasets, (ii) technical difficulties in adapting advanced deep neural networks to plant point clouds, and (iii) the lack of standardized benchmarks and evaluation protocols tailored to plant science. This review systematically addresses these barriers by: (i) providing an overview of existing 3D plant datasets in the context of general 3D segmentation domains, (ii) systematically summarizing deep learning-based methods for point cloud semantic and instance segmentation, (iii) introducing Plant Segmentation Studio (PSS), an open-source framework for reproducible benchmarking, and (iv) conducting extensive quantitative experiments to evaluate representative networks and sim-to-real learning strategies. Our findings highlight the efficacy of sparse-convolution and serialization-based backbones, as well as transformer-based instance segmentation networks, while also emphasizing the complementary role of modeling-based and augmentation-based synthetic data generation for sim-to-real learning in reducing annotation demands. Overall, this study bridges the gap between algorithmic advances and practical deployment, providing immediate tools for researchers and a roadmap for developing data-efficient and generalizable deep learning solutions in 3D plant phenotyping. Data and code are available at: https:\u002F\u002Fgithub.com\u002Fperrydoremi\u002FPlantSegStudio .","植物形态的精确表征为理解植物-环境相互作用和遗传演化提供了宝贵见解。提取此类信息的一项关键技术是三维分割，它能够从复杂点云中勾勒出植物各器官。尽管通用三维计算机视觉领域已取得显著进展，三维分割在植物表型分析中的应用仍受限于三大挑战：(i) 大规模标注数据集的稀缺，(ii) 将先进深度神经网络适配于植物点云的技术困难，以及(iii) 缺乏面向植物科学量身定制的标准化基准与评估协议。本综述通过以下方式系统性地应对这些障碍：(i) 在通用三维分割领域的背景下概述现有三维植物数据集，(ii) 系统总结基于深度学习的点云语义与实例分割方法，(iii) 介绍植物分割工作室（Plant Segmentation Studio, PSS）——一个用于可重复基准测试的开源框架，以及(iv) 开展大量定量实验以评估代表性网络和仿真到现实（sim-to-real）学习策略。我们的研究结果凸显了稀疏卷积与基于序列化的主干网络以及基于Transformer的实例分割网络的有效性，同时强调了基于建模和基于增强的合成数据生成在仿真到现实学习中对降低标注需求的互补作用。总体而言，本研究弥合了算法进展与实际部署之间的差距，为研究人员提供了即时可用的工具，并为在三维植物表型分析中开发数据高效且具有泛化能力的深度学习解决方案提供了路线图。数据与代码见：https:\u002F\u002Fgithub.com\u002Fperrydoremi\u002FPlantSegStudio 。",null,"ISPRS Journal of Photogrammetry and Remote Sensing","2026-09-16T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,9,1,"ISPRS顶刊综述，系统梳理器官级3D植物分割的数据-算法-算力瓶颈，并开源Plant Segmentation Studio基准框架，对作物表型与智慧育种有直接工具价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","开源工具","植物表型","三维点云",0,"10.1016\u002Fj.isprsjprs.2026.09.003",{"doi":32,"openalex_id":34,"authors":35,"venue":10,"cited_by_count":31,"oa_url":6,"card":60,"direction":64,"ingested_from":66},"W7213280479",[36,38,40,42,45,48,50,52,54,57],{"name":37,"orcid":9},"Ruiming Du",{"name":39,"orcid":9},"Guangxun Zhai",{"name":41,"orcid":9},"Tian Qiu",{"name":43,"orcid":44},"Shichao Jin","https:\u002F\u002Forcid.org\u002F0000-0003-1150-336X",{"name":46,"orcid":47},"Dawei Li","https:\u002F\u002Forcid.org\u002F0000-0002-9702-8848",{"name":49,"orcid":9},"Zhihong Ma",{"name":51,"orcid":9},"Yongliang Qiao",{"name":53,"orcid":9},"Junfeng Gao",{"name":55,"orcid":56},"Haiyan Cen","https:\u002F\u002Forcid.org\u002F0000-0003-0266-2123",{"name":58,"orcid":59},"Yu Jiang","https:\u002F\u002Forcid.org\u002F0000-0003-4495-3033",{"tldr":61,"method":62,"finding":63,"direction":64,"opportunity":65},"系统综述植物器官级3D分割，提出数据-算法-算力三角框架并开源基准平台PSS。","综述3D植物数据集与深度学习分割方法，构建PSS开源基准并做定量实验。","稀疏卷积与序列化骨干及Transformer实例分割有效，合成数据可降低标注需求。","农业遥感与作物表型","可探索数据高效、跨物种泛化的3D植物分割模型与标准化评测协议。","openalex","2026-09-17T23:30:30.425366Z"]