Low-Shot Learning of Plankton Categories.

Schröder, Simon-Martin , Kiko, Rainer , Irisson, Jean-Olivier and Koch, Reinhard (2019) Low-Shot Learning of Plankton Categories. In: Pattern Recognition - GCPR 2018. , ed. by Brox, T., Bruhn, A. and Fritz, M.. Lecture Notes in Computer Science, 11269 . Springer, Cham, Switzerland, pp. 391-404. ISBN 978-3-030-12939-2 DOI 10.1007/978-3-030-12939-2_27.

[thumbnail of 10.1007_978-3-030-12939-2_27.pdf] Text
10.1007_978-3-030-12939-2_27.pdf - Published Version
Restricted to Registered users only

Download (2MB) | Contact

Supplementary data:


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 re-training. This permits the mining of examples for novel classes.

Document Type: Book chapter
Research affiliation: OceanRep > SFB 754
OceanRep > The Future Ocean - Cluster of Excellence
OceanRep > GEOMAR > FB3 Marine Ecology > FB3-EOE-B Experimental Ecology - Benthic Ecology
Refereed: Yes
Publisher: Springer
Projects: SFB754, Future Ocean
Date Deposited: 10 Apr 2019 08:02
Last Modified: 23 Sep 2019 20:31
URI: https://oceanrep.geomar.de/id/eprint/46326

Actions (login required)

View Item View Item