PulseAugur
EN
LIVE 13:53:05

Planktonzilla dataset launches with 17.4M images for marine research

Researchers have introduced Planktonzilla-17M, a new dataset containing 17.4 million images of plankton, making it the largest of its kind. This dataset aims to improve plankton classification by consolidating images from thirteen different imaging systems and standardizing taxonomy and metadata. Experiments using Planktonzilla-17M showed that supervised classification with taxonomic lineage as text performed comparably to or better than CLIP-style image-text training, and highlighted limitations in current biological foundation models for marine imaging. AI

IMPACT Establishes a new benchmark for marine imaging AI, potentially improving ecological monitoring and climate modeling.

RANK_REASON The cluster contains a research paper detailing a new dataset and model for plankton classification.

Read on arXiv cs.NE (Neural & Evolutionary) →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.AI TIER_1 Dansk(DA) · Alan Gerson Contreras Montanares, Luis Valenzuela, Luis Mart\'i, Nayat Sanchez-Pi ·

    Planktonzilla: Multimodal dataset and models for understanding plankton ecosystems

    arXiv:2606.00080v1 Announce Type: cross Abstract: Marine plankton underpin aquatic food webs and play a key role in global CO2 sequestration, making reliable species identification critical for understanding ocean health and climate feedbacks. Existing classification models perfo…

  2. arXiv cs.NE (Neural & Evolutionary) TIER_1 Dansk(DA) · Nayat Sanchez-Pi ·

    Planktonzilla: Multimodal dataset and models for understanding plankton ecosystems

    Marine plankton underpin aquatic food webs and play a key role in global CO2 sequestration, making reliable species identification critical for understanding ocean health and climate feedbacks. Existing classification models perform well on individual collections but fail to gene…