Automatic photo orientation detection with convolutional neural networks

Ujash Joshi, Michael Guerzhoy

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

18 Scopus citations

Abstract

We apply convolutional neural networks (CNN) to the problem of image orientation detection in the context of determining the correct orientation (from 0, 90, 180, and 270 degrees) of a consumer photo. The problem is especially important for digitazing analog photographs. We substantially improve on the published state of the art in terms of the performance on one of the standard datasets, and test our system on a more difficult large dataset of consumer photos. We use Guided Backpropagation to obtain insights into how our CNN detects photo orientation, and to explain its mistakes.

Original languageEnglish (US)
Title of host publicationProceedings - 2017 14th Conference on Computer and Robot Vision, CRV 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages103-108
Number of pages6
ISBN (Electronic)9781538628188
DOIs
StatePublished - Jul 2 2017
Externally publishedYes
Event14th Conference on Computer and Robot Vision, CRV 2017 - Edmonton, Canada
Duration: May 17 2017May 19 2017

Publication series

NameProceedings - 2017 14th Conference on Computer and Robot Vision, CRV 2017
Volume2018-January

Conference

Conference14th Conference on Computer and Robot Vision, CRV 2017
Country/TerritoryCanada
CityEdmonton
Period5/17/175/19/17

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Keywords

  • convolutional neural networks
  • guided backpropagation
  • image orientation
  • photo
  • visualizing convnets

Fingerprint

Dive into the research topics of 'Automatic photo orientation detection with convolutional neural networks'. Together they form a unique fingerprint.

Cite this