Skip to main navigation Skip to search Skip to main content

DNN Flow: DNN feature pyramid based image matching

  • Wei Yu
  • , Kuiyuan Yang
  • , Yalong Bai
  • , Hongxun Yao
  • , Yong Rui
  • Harbin Institute of Technology
  • Microsoft USA

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Image matching especially in category level is a challenge but important problem in vision. The advance of image matching largely depends on the advance of image features. In viewing recent success of learned image feature by DNN, we propose an image matching algorithm based on DNN feature pyramid, named as DNN Flow. The nature of DNN feature pyramid in detecting different level patterns makes it is suitable to match two images in a coarse to fine manner, where top level coarsely matches two images in object level, middle level matches two images in part level, and low level finely matches two images in pixel level. The coarse to fine matching based on DNN feature pyramid is formulated as a series of optimization problems considering the guidance from top level. Extensive experiments demonstrate the superiority of DNN Flow in image matching under challenge variations.

Original languageEnglish
Title of host publicationBMVC 2014 25th British Machine Vision Conference 2014
EditorsMichel Valstar, Andrew French, Tony Pridmore
PublisherBritish Machine Vision Association, BMVA
ISBN (Print)1901725529
DOIs
StatePublished - 2014
Event25th British Machine Vision Conference, BMVC 2014 - Nottingham, United Kingdom
Duration: 1 Sep 20145 Sep 2014

Publication series

NameBMVC 2014 - Proceedings of the British Machine Vision Conference 2014

Conference

Conference25th British Machine Vision Conference, BMVC 2014
Country/TerritoryUnited Kingdom
CityNottingham
Period1/09/145/09/14

Fingerprint

Dive into the research topics of 'DNN Flow: DNN feature pyramid based image matching'. Together they form a unique fingerprint.

Cite this