A Multi-class Kernel Alignment Method for Image Collection Summarization

Authors:
Jorge E. Camargo;Fabio A. González
Affiliations:
Bioingenium Research Group, National University of Colombia,;Bioingenium Research Group, National University of Colombia,
Venue:
CIARP '09 Proceedings of the 14th Iberoamerican Conference on Pattern Recognition: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
Year:
2009

Citing 9
Cited 0

eID: a system for exploration of image databases

Information Processing and Management: an International Journal
Kernel Methods for Pattern Analysis

Kernel Methods for Pattern Analysis
Hierarchical clustering of WWW image search results using visual, textual and link information

Proceedings of the 12th annual ACM international conference on Multimedia
Introduction to Machine Learning (Adaptive Computation and Machine Learning)

Introduction to Machine Learning (Adaptive Computation and Machine Learning)
Web image clustering by consistent utilization of visual features and surrounding texts

Proceedings of the 13th annual ACM international conference on Multimedia
Content-based image collection summarization and comparison using self-organizing maps

Pattern Recognition
Image retrieval: Ideas, influences, and trends of the new age

ACM Computing Surveys (CSUR)
Interactive access to large image collections using similarity-based visualization

Journal of Visual Languages and Computing
Hierarchical browsing and search of large image databases

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a method for involving domain knowledge in the construction of summaries of large collections of images. This is accomplished by using a multi-class kernel alignment strategy in order to learn a kernel function that incorporates domain knowledge (class labels). The kernel function is the basis of a clustering algorithm that generates a subset, the summary, of the image collection. The method was tested with a subset of the Corel image collection using a summarization quality measure based on information theory. Experimental results show that it is possible to improve the quality of the summary when domain knowledge is involved.