PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation

Liu, Xinpeng; Santo, Hiroaki; Toda, Yosuke; Okura, Fumio

doi:https://doi.org/10.1007/s11263-026-02882-4

PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation

Liu, Xinpeng; Santo, Hiroaki; Toda, Yosuke et al.

International Journal of Computer Vision, 2026, 134, 314

アクセス数:221件（2026-10-25 01:23 集計）

固定URL: https://hdl.handle.net/11094/105906

閲覧可能ファイル

ファイル	フォーマット	利用条件	サイズ	閲覧回数	利用開始日	説明	information
IntJComputVision_134_314	pdf	なし	28.2 MB	48	2026.07.30

Creative Commons : 表示

論文情報

ファイル出力 EndNote Basic出力 Mendeley出力

タイトル	PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation
著者	著者名 Liu, Xinpeng
	NRID 1000040830906 1000040830906 ORCID 0000-0003-2891-5993 0000-0003-2891-5993 researchmap permalink h_santo h_santo Scopus Author ID 57194410230 57194410230 著者名 Santo, Hiroaki
	NRID 1000000808264 1000000808264 著者名 Toda, Yosuke
	NRID 1000060754223 1000060754223 ORCID 0000-0001-7595-1300 0000-0001-7595-1300 researchmap permalink okura.fumio okura.fumio Scopus Author ID 57212779493 57212779493 著者名 Okura, Fumio
キーワード等	Agriculture
	Image-to-graph generation
	Plant phenotyping
内容	Liu X., Santo H., Toda Y., et al. PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation. International Journal of Computer Vision 134, 314 (2026); https://doi.org/10.1007/s11263-026-02882-4.
抄録	Accurate estimation of plant skeletal structures (e.g., branching structures) from images is essential for smart agriculture and plant science. Unlike human skeletons with fixed topology, plant skeleton estimation presents a unique challenge, i.e., estimating arbitrary tree graphs from images. To address this problem, we introduce PlantPose, a universal plant skeleton estimator via tree-constrained graph generation. PlantPose combines learning-based graph generation with traditional graph algorithms to enforce tree constraints during the training loop. To enhance the model’s generalization capability, we curate a large and diverse dataset comprising real-world and synthetic plant images, along with simplified representations (e.g., sketches and abstract drawings). This dataset enables the generalized model to adapt to diverse input styles and categories of plant images while preserving topological consistency. Our approach demonstrates robust and accurate plant skeleton estimation across multiple domains, including previously unseen out-of-domain scenarios. Further analyses highlight the method’s strengths and limitations in handling complex, heterogeneous data distributions. All implementations and datasets are available at https://github.com/huntorochi/PlantPose/.
出版者	Springer
収録物名	International Journal of Computer Vision
巻(号)	134
ページ	314
刊行年月	2026-06-10
言語	英語
ハンドルURL	https://hdl.handle.net/11094/105906
PISSN	09205691
eISSN	15731405
関連情報 (isIdenticalTo)	DOI (出版社版) https://doi.org/10.1007/s11263-026-02882-4
アクセス権	open access
権利情報	This article is licensed under a Creative Commons Attribution 4.0 International License.
出版タイプ	Version of Record
カテゴリ	学術雑誌論文 Journal Article

資源タイプ	学術雑誌論文
ローカル資源タイプ	学術雑誌論文
DCMI資源タイプ	text
DCTERMS.bibliographicCitation	International Journal of Computer Vision. 2026, 134, p. 314
DC.title	PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation
DC.creator	Liu, Xinpeng
	Santo, Hiroaki
	Toda, Yosuke
	Okura, Fumio
DC.publisher	Springer
DC.language' scheme='DCTERMS.RFC1766	英語
DC.type' scheme='DCTERMS.DCMIType	text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2026-06-10
DC.identifier	https://doi.org/10.1007/s11263-026-02882-4
DC.subject	Agriculture
	Image-to-graph generation
	Plant phenotyping
DC.description	Liu X., Santo H., Toda Y., et al. PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation. International Journal of Computer Vision 134, 314 (2026); https://doi.org/10.1007/s11263-026-02882-4.
DCTERMS.abstract	Accurate estimation of plant skeletal structures (e.g., branching structures) from images is essential for smart agriculture and plant science. Unlike human skeletons with fixed topology, plant skeleton estimation presents a unique challenge, i.e., estimating arbitrary tree graphs from images. To address this problem, we introduce PlantPose, a universal plant skeleton estimator via tree-constrained graph generation. PlantPose combines learning-based graph generation with traditional graph algorithms to enforce tree constraints during the training loop. To enhance the model’s generalization capability, we curate a large and diverse dataset comprising real-world and synthetic plant images, along with simplified representations (e.g., sketches and abstract drawings). This dataset enables the generalized model to adapt to diverse input styles and categories of plant images while preserving topological consistency. Our approach demonstrates robust and accurate plant skeleton estimation across multiple domains, including previously unseen out-of-domain scenarios. Further analyses highlight the method’s strengths and limitations in handling complex, heterogeneous data distributions. All implementations and datasets are available at https://github.com/huntorochi/PlantPose/.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	105906
citation_title	PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation
citation_author	Liu, Xinpeng
	Santo, Hiroaki
	Toda, Yosuke
	Okura, Fumio
citation_publisher	Springer
citation_language	英語
citation_date	2026-06-10
citation_journal_title	International Journal of Computer Vision
citation_volume	134
citation_firstpage	314
citation_doi	https://doi.org/10.1007/s11263-026-02882-4
citation_public_url	https://hdl.handle.net/11094/105906

PlantPose: Universal Plant Skeleton Estimation via Tree-constrained Graph Generation

閲覧可能ファイル

論文情報

論文詳細を表示