Maximum classification optimization-based active learning for image classification

Zhengwei Cui, Xiaoming Chen, Jian Wu, Victor S. Sheng, Yujie Shi

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

3 Scopus citations

Abstract

Traditional multi-class image classification needs a large number of training samples for building a classifier model. However, it is very time-consuming and costly to obtain labels for a large number of training samples from human experts. Active learning is a feasible solution. This paper proposes a maximum classification optimization method (MCO) for actively selecting unlabeled images to acquire labels. It integrated the information of an unlabeled sample from different perspectives with two steps. It first chooses a subset of candidates, and then selects the best from these candidates. Our experimental results show that the maximum classification optimization method outperforms two popular exiting methods (entropy-based uncertainty and BvSB).

Original languageEnglish
Title of host publicationProceedings - 2014 7th International Congress on Image and Signal Processing, CISP 2014
EditorsYi Wan, Jinguang Sun, Jingchang Nan, Quangui Zhang, Liangshan Shao, Lipo Wang
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages759-764
Number of pages6
ISBN (Electronic)9781479958351
DOIs
StatePublished - Jan 6 2014
Event2014 7th International Congress on Image and Signal Processing, CISP 2014 - Dalian, China
Duration: Oct 14 2014Oct 16 2014

Publication series

NameProceedings - 2014 7th International Congress on Image and Signal Processing, CISP 2014

Conference

Conference2014 7th International Congress on Image and Signal Processing, CISP 2014
CountryChina
CityDalian
Period10/14/1410/16/14

Keywords

  • BvSB
  • active learning
  • image classification
  • maximum classification optimization

Fingerprint Dive into the research topics of 'Maximum classification optimization-based active learning for image classification'. Together they form a unique fingerprint.

Cite this