Visual tracking by the combination of global detector and local image patch matching

Li Sun, Kai Qu, Shanshan Xu, Song Qiu

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

Abstract

This paper presents an approach for visual tracking, consisting of two combination modules, which are global detector and local image patch matching. The former gives the classification response for each object candidate specified by the sliding window in the searching region. The classification can be performed by any global detector, which is based on the feature from the local patch in the object region. To cope with the pose change or deformation of object, the local patch in the object region is allowed to drift in two adjacent frames by searching the best matching position and quantifying the matching metric. Experiments demonstrate the performance of the algorithm especially under certain special cases.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages3474-3478
Number of pages5
ISBN (Electronic)9781467399616
DOIs
StatePublished - 3 Aug 2016
Event23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States
Duration: 25 Sep 201628 Sep 2016

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2016-August
ISSN (Print)1522-4880

Conference

Conference23rd IEEE International Conference on Image Processing, ICIP 2016
Country/TerritoryUnited States
CityPhoenix
Period25/09/1628/09/16

Keywords

  • Detection
  • Template matching
  • Tracking

Fingerprint

Dive into the research topics of 'Visual tracking by the combination of global detector and local image patch matching'. Together they form a unique fingerprint.

Cite this