Low-shot learning of plankton categories - CNRS - Centre national de la recherche scientifique Access content directly
Proceedings Year : 2019

Low-shot learning of plankton categories

Simon-Martin Schröder
Rainer Kiko
Reinhard Koch

Abstract

The size of current plankton image datasets renders manual classification virtually infeasible. The training of models for machine classification is complicated by the fact that a large number of classes consist of only a few examples. We employ the recently introduced weight imprinting technique in order to use the available training data to train accurate classifiers in absence of enough examples for some classes. The model architecture used in this work succeeds in the identification of plankton using machine learning with its unique challenges, i.e. a limited number of training examples and a severely skewed class size distribution. Weight imprinting enables a neural network to recognize small classes immediately without retraining. This permits the mining of examples for novel classes.
Fichier principal
Vignette du fichier
Schroder2018-Low-shot learning of plankton categories00.pdf (3.16 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03390570 , version 1 (21-10-2021)

Identifiers

Cite

Simon-Martin Schröder, Rainer Kiko, Jean-Olivier Irisson, Reinhard Koch. Low-shot learning of plankton categories. 11269, 2019, 978-3-030-12939-2. ⟨10.1007/978-3-030-12939-2_27⟩. ⟨hal-03390570⟩
19 View
43 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More