Image classification with recurrent attention models

Stanislau Semeniuta, Erhardt Barth

Abstract

In this work we apply a fully differentiable Recurrent Model of Visual Attention to unconstrained real-world images. We propose a deep recurrent attention model and show that it can successfully learn to jointly localize and classify objects. We evaluate our model on multiple digit images generated from MNIST data, Google Street View images, and a fine-grained recognition dataset of 200 bird species, and show that its performance is either comparable or superior to that of alternative models.

Original languageEnglish
Title of host publication2016 IEEE Symposium Series on Computational Intelligence (SSCI)
PublisherIEEE
Publication date09.02.2017
Article number7850113
ISBN (Print)978-1-5090-4241-8
ISBN (Electronic)978-1-5090-4240-1
DOIs
Publication statusPublished - 09.02.2017
Event2016 IEEE Symposium Series on Computational Intelligence - Athens, Greece
Duration: 06.12.201609.12.2016
Conference number: 126460

Fingerprint

Dive into the research topics of 'Image classification with recurrent attention models'. Together they form a unique fingerprint.

Cite this