3DPotatoTwin: a paired potato tuber dataset for 3D multi-sensory fusion. Wang, H., Blok, P. M., Burridge, J., Jiang, T., Miyauchi, M., Miyamoto, K., Tanaka, K., & Guo, W. Plant Phenomics, 7(4):100123, December, 2025. ![3DPotatoTwin: a paired potato tuber dataset for 3D multi-sensory fusion [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable highthroughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin, a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.
@article{wang_3dpotatotwin_2025,
title = {{3DPotatoTwin}: a paired potato tuber dataset for {3D} multi-sensory fusion},
volume = {7},
issn = {26436515},
shorttitle = {{3DPotatoTwin}},
url = {https://linkinghub.elsevier.com/retrieve/pii/S2643651525001293},
doi = {10.1016/j.plaphe.2025.100123},
abstract = {Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable highthroughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin, a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.},
language = {en},
number = {4},
urldate = {2025-12-14},
journal = {Plant Phenomics},
author = {Wang, Haozhou and Blok, Pieter M. and Burridge, James and Jiang, Ting and Miyauchi, Minato and Miyamoto, Kyosuke and Tanaka, Kunihiro and Guo, Wei},
month = dec,
year = {2025},
pages = {100123},
}
Downloads: 0
{"_id":"2zAs5Wd37epzTCJ9H","bibbaseid":"wang-blok-burridge-jiang-miyauchi-miyamoto-tanaka-guo-3dpotatotwinapairedpotatotuberdatasetfor3dmultisensoryfusion-2025","author_short":["Wang, H.","Blok, P. M.","Burridge, J.","Jiang, T.","Miyauchi, M.","Miyamoto, K.","Tanaka, K.","Guo, W."],"bibdata":{"bibtype":"article","type":"article","title":"3DPotatoTwin: a paired potato tuber dataset for 3D multi-sensory fusion","volume":"7","issn":"26436515","shorttitle":"3DPotatoTwin","url":"https://linkinghub.elsevier.com/retrieve/pii/S2643651525001293","doi":"10.1016/j.plaphe.2025.100123","abstract":"Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable highthroughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin, a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.","language":"en","number":"4","urldate":"2025-12-14","journal":"Plant Phenomics","author":[{"propositions":[],"lastnames":["Wang"],"firstnames":["Haozhou"],"suffixes":[]},{"propositions":[],"lastnames":["Blok"],"firstnames":["Pieter","M."],"suffixes":[]},{"propositions":[],"lastnames":["Burridge"],"firstnames":["James"],"suffixes":[]},{"propositions":[],"lastnames":["Jiang"],"firstnames":["Ting"],"suffixes":[]},{"propositions":[],"lastnames":["Miyauchi"],"firstnames":["Minato"],"suffixes":[]},{"propositions":[],"lastnames":["Miyamoto"],"firstnames":["Kyosuke"],"suffixes":[]},{"propositions":[],"lastnames":["Tanaka"],"firstnames":["Kunihiro"],"suffixes":[]},{"propositions":[],"lastnames":["Guo"],"firstnames":["Wei"],"suffixes":[]}],"month":"December","year":"2025","pages":"100123","bibtex":"@article{wang_3dpotatotwin_2025,\n\ttitle = {{3DPotatoTwin}: a paired potato tuber dataset for {3D} multi-sensory fusion},\n\tvolume = {7},\n\tissn = {26436515},\n\tshorttitle = {{3DPotatoTwin}},\n\turl = {https://linkinghub.elsevier.com/retrieve/pii/S2643651525001293},\n\tdoi = {10.1016/j.plaphe.2025.100123},\n\tabstract = {Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable highthroughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin, a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.},\n\tlanguage = {en},\n\tnumber = {4},\n\turldate = {2025-12-14},\n\tjournal = {Plant Phenomics},\n\tauthor = {Wang, Haozhou and Blok, Pieter M. and Burridge, James and Jiang, Ting and Miyauchi, Minato and Miyamoto, Kyosuke and Tanaka, Kunihiro and Guo, Wei},\n\tmonth = dec,\n\tyear = {2025},\n\tpages = {100123},\n}\n\n\n\n\n\n\n\n\n\n\n\n","author_short":["Wang, H.","Blok, P. M.","Burridge, J.","Jiang, T.","Miyauchi, M.","Miyamoto, K.","Tanaka, K.","Guo, W."],"key":"wang_3dpotatotwin_2025","id":"wang_3dpotatotwin_2025","bibbaseid":"wang-blok-burridge-jiang-miyauchi-miyamoto-tanaka-guo-3dpotatotwinapairedpotatotuberdatasetfor3dmultisensoryfusion-2025","role":"author","urls":{"Paper":"https://linkinghub.elsevier.com/retrieve/pii/S2643651525001293"},"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://bibbase.org/zotero-group/howcanoewang/4660314","dataSources":["AHMB37rGQdtcwer6D","z5vjAHmuibbBBnvdc"],"keywords":[],"search_terms":["3dpotatotwin","paired","potato","tuber","dataset","multi","sensory","fusion","wang","blok","burridge","jiang","miyauchi","miyamoto","tanaka","guo"],"title":"3DPotatoTwin: a paired potato tuber dataset for 3D multi-sensory fusion","year":2025}