Design of kernels in convolutional neural networks for image classification

Zhun Sun, Mete Ozay, Takayuki Okatani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

Despite the effectiveness of convolutional neural networks (CNNs) for image classification, our understanding of the effect of shape of convolution kernels on learned representations is limited. In this work, we explore and employ the relationship between shape of kernels which define receptive fields (RFs) in CNNs for learning of feature representations and image classification. For this purpose, we present a feature visualization method for visualization of pixel-wise classification score maps of learned features. Motivated by our experimental results, and observations reported in the literature for modeling of visual systems, we propose a novel design of shape of kernels for learning of representations in CNNs. In the experimental results, the proposed models also outperform the state-of-the-art methods employed on the CIFAR-10/100 datasets [1] for image classification. We also achieved an outstanding performance in the classification task, comparing to a base CNN model that introduces more parameters and computational time, using the ILSVRC-2012 dataset [2]. Additionally, we examined the region of interest (ROI) of different models in the classification task and analyzed the robustness of the proposed method to occluded images. Our results indicate the effectiveness of the proposed approach.

Original languageEnglish
Title of host publicationComputer Vision - 14th European Conference, ECCV 2016, Proceedings
EditorsBastian Leibe, Jiri Matas, Nicu Sebe, Max Welling
PublisherSpringer Verlag
Pages51-66
Number of pages16
ISBN (Print)9783319464770
DOIs
Publication statusPublished - 2016
Event14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands
Duration: 2016 Oct 82016 Oct 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9911 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th European Conference on Computer Vision, ECCV 2016
CountryNetherlands
CityAmsterdam
Period16/10/816/10/16

Keywords

  • Convolution kernel
  • Convolutional neural networks
  • Deep learning
  • Image classification
  • Kernel design

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Design of kernels in convolutional neural networks for image classification'. Together they form a unique fingerprint.

Cite this