Saved in:
Bibliographic Details
Main Authors: Chen, Yueqianji, Williams, Kevin, Doonan, John H., Remagnino, Paolo, Hepworth, Jo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.15445
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917151833587712
author Chen, Yueqianji
Williams, Kevin
Doonan, John H.
Remagnino, Paolo
Hepworth, Jo
author_facet Chen, Yueqianji
Williams, Kevin
Doonan, John H.
Remagnino, Paolo
Hepworth, Jo
contents Automated extraction of individual plant branches from time-series imagery is essential for high-throughput phenotyping, yet it remains computationally challenging due to non-rigid growth dynamics and severe identity fragmentation within entangled canopies. To overcome these stage-dependent ambiguities, we propose ST-DETrack, a spatiotemporal-fusion dual-decoder network designed to preserve branch identity from budding to flowering. Our architecture integrates a spatial decoder, which leverages geometric priors such as position and angle for early-stage tracking, with a temporal decoder that exploits motion consistency to resolve late-stage occlusions. Crucially, an adaptive gating mechanism dynamically shifts reliance between these spatial and temporal cues, while a biological constraint based on negative gravitropism mitigates vertical growth ambiguities. Validated on a Brassica napus dataset, ST-DETrack achieves a Branch Matching Accuracy (BMA) of 93.6%, significantly outperforming spatial and temporal baselines by 28.9 and 3.3 percentage points, respectively. These results demonstrate the method's robustness in maintaining long-term identity consistency amidst complex, dynamic plant architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2512_15445
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ST-DETrack: Identity-Preserving Branch Tracking in Entangled Plant Canopies via Dual Spatiotemporal Evidence
Chen, Yueqianji
Williams, Kevin
Doonan, John H.
Remagnino, Paolo
Hepworth, Jo
Computer Vision and Pattern Recognition
Automated extraction of individual plant branches from time-series imagery is essential for high-throughput phenotyping, yet it remains computationally challenging due to non-rigid growth dynamics and severe identity fragmentation within entangled canopies. To overcome these stage-dependent ambiguities, we propose ST-DETrack, a spatiotemporal-fusion dual-decoder network designed to preserve branch identity from budding to flowering. Our architecture integrates a spatial decoder, which leverages geometric priors such as position and angle for early-stage tracking, with a temporal decoder that exploits motion consistency to resolve late-stage occlusions. Crucially, an adaptive gating mechanism dynamically shifts reliance between these spatial and temporal cues, while a biological constraint based on negative gravitropism mitigates vertical growth ambiguities. Validated on a Brassica napus dataset, ST-DETrack achieves a Branch Matching Accuracy (BMA) of 93.6%, significantly outperforming spatial and temporal baselines by 28.9 and 3.3 percentage points, respectively. These results demonstrate the method's robustness in maintaining long-term identity consistency amidst complex, dynamic plant architectures.
title ST-DETrack: Identity-Preserving Branch Tracking in Entangled Plant Canopies via Dual Spatiotemporal Evidence
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.15445