TY - DATA T1 - TomatoWUR: an annotated dataset of 3D tomato plants to quantitatively evaluate segmentation, skeletonisation, and plant trait extraction algorithms for 3D plant phenotyping PY - 2025/03/27 AU - Bart M. Marrewijk, van AU - Bolai Xin AU - Tim van Daalen AU - Eldert van Henten AU - Gerrit Polder AU - Gert Kootstra UR - DO - 10.4121/e2c59841-4653-45de-a75e-4994b2766a2f.v1 KW - 3D plant phenotyping KW - horticulture KW - tomato KW - segmentation KW - plant architecture KW - point clouds KW - RGB images KW - 3D phenotyping KW - shape-from-silhouette N2 -

In this research, we created a dataset of 44 annotated 3D point clouds of tomato plants. Images were captured using fifteen cameras surrounding a single plant. Those images were used to create a point cloud using the shape-form silhouette methodology (Koenderink et al., 2009). The resulting dataset was annotated and separated into three parts. 1) Annotated point clouds with semantic and instance labels. The annotated point clouds were used to render annotated RGB images. 2) Annotated skeletons to analyse plant architecture. 3) Manual measurements of internode length, stem thickness, branching and phyllotactic angle to evaluate phenotyping algorithms from point cloud to plant traits. To view the dataset, please have a look at our git:

ER -