Dual-Margin Embedding for Fine-Grained Long-Tailed Plant Taxonomy

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Low, Cheng Yaw, Koo, Heejoon, Park, Jaewoo, Cha, Meeyoung
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913025666056192
author Low, Cheng Yaw
Koo, Heejoon
Park, Jaewoo
Cha, Meeyoung
author_facet Low, Cheng Yaw
Koo, Heejoon
Park, Jaewoo
Cha, Meeyoung
contents Taxonomic classification of ecological families, genera, and species underpins biodiversity monitoring and conservation. Existing computer vision methods typically address fine-grained recognition and long-tailed learning in isolation. However, additional challenges such as spatiotemporal domain shift, hierarchical taxonomic structure, and previously unseen taxa often co-occur in real-world deployment, leading to brittle performance under open-world conditions. We propose TaxoNet, an embedding learning framework with a theoretically grounded dual-margin objective that reshapes class decision boundaries under class imbalance to improve fine-grained discrimination while strengthening rare-class representation geometry. We evaluate TaxoNet in open-world settings that capture co-occurring recognition challenges. Leveraging diverse plant datasets, including Google Auto-Arborist (urban tree imagery), iNaturalist (Plantae observations across heterogeneous ecosystems), and NAFlora-Mini (herbarium collections), we demonstrate that TaxoNet consistently outperforms strong baselines, including multimodal foundation models.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18994
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual-Margin Embedding for Fine-Grained Long-Tailed Plant Taxonomy
Low, Cheng Yaw
Koo, Heejoon
Park, Jaewoo
Cha, Meeyoung
Computer Vision and Pattern Recognition
Taxonomic classification of ecological families, genera, and species underpins biodiversity monitoring and conservation. Existing computer vision methods typically address fine-grained recognition and long-tailed learning in isolation. However, additional challenges such as spatiotemporal domain shift, hierarchical taxonomic structure, and previously unseen taxa often co-occur in real-world deployment, leading to brittle performance under open-world conditions. We propose TaxoNet, an embedding learning framework with a theoretically grounded dual-margin objective that reshapes class decision boundaries under class imbalance to improve fine-grained discrimination while strengthening rare-class representation geometry. We evaluate TaxoNet in open-world settings that capture co-occurring recognition challenges. Leveraging diverse plant datasets, including Google Auto-Arborist (urban tree imagery), iNaturalist (Plantae observations across heterogeneous ecosystems), and NAFlora-Mini (herbarium collections), we demonstrate that TaxoNet consistently outperforms strong baselines, including multimodal foundation models.
title Dual-Margin Embedding for Fine-Grained Long-Tailed Plant Taxonomy
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.18994