BioCosmos: computer vision and large language models for large-scale image-based biodiversity data and trait exploration
Abstract
Millions of natural history specimen images are now online, but poor discoverability keeps their rich phenotypic and evolutionary data out of reach for most users. We developed BioCosmos, a web platform that combines conventional biodiversity databases with emerging computer vision and large language models for image retrieval, visual feature recognition, and natural language queries. Its first implementation, LepiVerse (https://lepiverse.com), is a butterfly web portal with more than 600,000 specimen images aggregated from natural history museums worldwide via GBIF, SCAN, and Ecdysis, and curated through taxonomic harmonization and georeference validation pipelines. BioCosmos features visual similarity, interactive morphospace plots, and text- and image-based searches, alongside species pages enriched with data from GenBank, LepTraits, Wikipedia, and Crossref. The platform demonstrates how an open-source, low-cost emerging AI can make vast image collections accessible to a broad audience, while highlighting the digitization gap between the Global North and South.
Related
- Design and implementation of an automated Drosophila locomotor assay using computer vision tracking
- Computer vision-aided locomotor behavioral analysis identifies therapeutic motor signatures in a mouse model of Huntington disease
- Impact Of Honey Bee Waggle Dance Recording Framerate On Computer Vision Process
- A modular open-source computer vision pipeline enables individual-level behavior analysis of group-housed pigs
- A Computer Vision-Based Framework for Multilevel Egg Fertility Classification Using L*a*b* Color Segmentation, Wavelet Texture Features, and Support Vector Machine
BibTeX
@misc{handika2026biocosmos,
title = {{BioCosmos: computer vision and large language models for large-scale image-based biodiversity data and trait exploration}},
author = {Heru Handika and Moritz D. Lürig and Kira L. Nichtawitz and Oliver M. Dobon and Venkata S. Yarava and Justin K. Sobayo and Breanna Shi and Michael J. Elliott and Jose A. B. Fortes and Arthur Porto},
howpublished = {EcoEvoRxiv},
year = {2026},
month = nov,
doi = {10.32942/x2c11r},
url = {https://ecoevorxiv.org/repository/view/15964/},
note = {CC BY-NC 4.0},
}
RIS
TY - UNPB
TI - BioCosmos: computer vision and large language models for large-scale image-based biodiversity data and trait exploration
AU - Heru Handika
AU - Moritz D. Lürig
AU - Kira L. Nichtawitz
AU - Oliver M. Dobon
AU - Venkata S. Yarava
AU - Justin K. Sobayo
AU - Breanna Shi
AU - Michael J. Elliott
AU - Jose A. B. Fortes
AU - Arthur Porto
PY - 2026
DA - 2026/11/05
PB - EcoEvoRxiv
DO - 10.32942/x2c11r
UR - https://ecoevorxiv.org/repository/view/15964/
AB - Millions of natural history specimen images are now online, but poor discoverability keeps their rich phenotypic and evolutionary data out of reach for most users. We developed BioCosmos, a web platform that combines conventional biodiversity databases with emerging computer vision and large language models for image retrieval, visual feature recognition, and natural language queries. Its first implementation, LepiVerse (https://lepiverse.com), is a butterfly web portal with more than 600,000 specimen images aggregated from natural history museums worldwide via GBIF, SCAN, and Ecdysis, and curated through taxonomic harmonization and georeference validation pipelines. BioCosmos features visual similarity, interactive morphospace plots, and text- and image-based searches, alongside species pages enriched with data from GenBank, LepTraits, Wikipedia, and Crossref. The platform demonstrates how an open-source, low-cost emerging AI can make vast image collections accessible to a broad audience, while highlighting the digitization gap between the Global North and South.
ER -