Cascade of multi-level multi-instance classifiers for image annotation

Cam Tu Nguyen, Ha Vu Le, Takeshi Tokuyama

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

3 Citations (Scopus)

Abstract

This paper introduces a new scheme for automatic image annotation based on cascading multi-level multi-instance classifiers (CMLMI). The proposed scheme employs a hierarchy for visual feature extraction, in which the feature set includes features extracted from the whole image at the coarsest level and from the overlapping sub-regions at finer levels. Multi-instance learning (MIL) is used to learn the "weak classifiers" for these levels in a cascade manner. The underlying idea is that the coarse levels are suitable for background labels such as "forest" and "city", while finer levels bring useful information about foreground objects like "tiger" and "car". The cascade manner allows this scheme to incorporate "important" negative samples during the learning process, hence reducing the "weakly labeling" problem by excluding ambiguous background labels associated with the negative samples. Experiments show that the CMLMI achieve significant improvements over baseline methods as well as existing MIL-based methods.

Original languageEnglish
Title of host publicationKDIR 2011 - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval
Pages14-23
Number of pages10
Publication statusPublished - 2011 Dec 1
EventInternational Conference on Knowledge Discovery and Information Retrieval, KDIR 2011 - Paris, France
Duration: 2011 Oct 262011 Oct 29

Publication series

NameKDIR 2011 - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval

Other

OtherInternational Conference on Knowledge Discovery and Information Retrieval, KDIR 2011
CountryFrance
CityParis
Period11/10/2611/10/29

Keywords

  • Cascade algorithm
  • Image annotation
  • Multi-level feature extraction

ASJC Scopus subject areas

  • Information Systems

Fingerprint Dive into the research topics of 'Cascade of multi-level multi-instance classifiers for image annotation'. Together they form a unique fingerprint.

Cite this