Deep learning for logo recognition

In this paper we propose a method for logo recognition using deep learning. Our recognition pipeline is composed of a logo region proposal followed by a Convolutional Neural Network (CNN) specifically trained for logo classification, even if they are not precisely localized. Experiments are carried...

Full description

Saved in:
Bibliographic Details
Published inNeurocomputing (Amsterdam) Vol. 245; pp. 23 - 30
Main Authors Bianco, Simone, Buzzelli, Marco, Mazzini, Davide, Schettini, Raimondo
Format Journal Article
LanguageEnglish
Published Elsevier B.V 05.07.2017
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:In this paper we propose a method for logo recognition using deep learning. Our recognition pipeline is composed of a logo region proposal followed by a Convolutional Neural Network (CNN) specifically trained for logo classification, even if they are not precisely localized. Experiments are carried out on the FlickrLogos-32 database, and we evaluate the effect on recognition performance of synthetic versus real data augmentation, and image pre-processing. Moreover, we systematically investigate the benefits of different training choices such as class-balancing, sample-weighting and explicit modeling the background class (i.e. no-logo regions). Experimental results confirm the feasibility of the proposed method, that outperforms the methods in the state of the art.
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2017.03.051