Data and Models for: IsPlanktonBIO

Read full text Summarise with VIRAT
Authors: David Sosa Trejo, Antonio J. Bandera, Martín González, Santiago Hernández‐León
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This repository contains the pre-trained model weights, vector databases (FAISS), and SQL relational databases required to run the IsPlanktonBIO framework, for both the desktop/GPU pipeline and its AMD Kria KV260 edge deployment. IsPlanktonBIO is an end-to-end pipeline designed for the automatic identification and biomass estimation of in situ plankton images. Due to file size limitations, these heavy assets are hosted here on Zenodo and are intended to be used alongside the source code at GitHub. Contents: models/weights/: Pre-trained ResNet50 PyTorch weights (Stage-1 and Stage-2) for the desktop/GPU pipeline. kria/models/: DPU-compiled INT8 ResNet50 weights (.xmodel, Vitis AI) for the AMD Kria KV260 edge deployment (Stage-1 and Stage-2). data/databases/vector/: FAISS indexes for fast similarity search. data/databases/sql/: SQLite databases containing biological taxonomy mapping.

Fields of Study & Subjects

Advanced Neural Network Applications Water Quality Monitoring Technologies Domain Adaptation and Few-Shot Learning

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
22 March 2026
Language
EN
Document type
Dataset