Naturalis Multi-Source Model 2023 Release
Zenodo (CERN European Organization for Nuclear Research)
Abstract
This repository contains the MSM23 (multi-source model 2023) model release package, which includes the trained models, associated metadata and installation instructions for the MSM23 project. The model identifies species of plants, animals and fungi that naturally occur in Europe, based on photos of observations. The underlying recognition model has been trained using photos from Observation International, Artsdatabanken (Norway), Artdatabanken (Sweden), United Kingdom Species Inventory (United Kingdom), Arter.dk (Denmark) and FINBIF (Finland). The model contains 39477 taxa: 31419 species, 3391 higher taxa, 949 species-groups, 3059 infraspecies and 659 hybrid species. Identifications can benefit from multiple photos from the same observation. Provide up to four photos in a single request to improve identification results. The model identifies one species per photo and one species per observation.
Authors 5
-
Affiliation as printed
Naturalis Biodiversity Center
-
Affiliation as printed
Naturalis Biodiversity Center
-
Affiliation as printed
Naturalis Biodiversity Center
-
Affiliation as printed
Naturalis Biodiversity Center
-
Affiliation as printed
Naturalis Biodiversity Center
Cited by 0 stored of 0
No patents citing this paper on Lens.org (checked 2026-10-11).