Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: A generalized probabilistic framework for compact codebook creation
Author: Liu, L.
Wang, L.
Shen, C.
Citation: IEEE CVPR 2011 Conference Colorado Springs: Computer Vision and Pattern Recognition (CVPR) 2011, June 21-23, 2011. 9p.
Publisher: IEEE
Publisher Place: USA
Issue Date: 2011
Series/Report no.: IEEE Conference on Computer Vision and Pattern Recognition
ISBN: 9781457703942
ISSN: 1063-6919
Conference Name: Computer Vision and Pattern Recognition (2011 : Colorado Springs, US)
Statement of
Lingqiao Liu, Lei Wang and Chunhua Shen
Abstract: Compact and discriminative visual codebooks are pre-ferred in many visual recognition tasks. In the literature, a few researchers have taken the approach of hierarchically merging visual words of a initial large-size code-book, but implemented this idea with different merging cri- teria. In this work, we show that by defining different class-conditional distribution functions and parameter estimation methods, these merging criteria can be unified under a single probabilistic framework. More importantly, by adopting new distribution functions and/or parameter estimation methods, we can generalize this framework to produce a spectrum of novel merging criteria. Two of them are particularly focused in this work. For one criterion, we adopt the multinomial distribution to model each object class, and for the other criterion we propose a large-margin based parameter estimation method. Both theoretical analysis and experimental study demonstrate the superior performance of the two new merging criteria and the general applicability of our probabilistic framework.
Description: Appearing in IEEE Conf. Comp. Vis. Pattern Recogn. 2011. This reprint differs from the original in pagination and typographic detail
Rights: © 2011 IEEE
DOI: 10.1109/CVPR.2011.5995628
Grant ID:
Description (link):
Published version:
Appears in Collections:Aurora harvest
Computer Science publications

Files in This Item:
File Description SizeFormat 
hdl_63014.pdfAccepted version358.1 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.