@ARTICLE{5128909,
author={J. C. van Gemert and C. J. Veenman and A. W. M. Smeulders and J. M. Geusebroek},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
title={Visual Word Ambiguity},
year={2010},
volume={32},
number={7},
pages={1271-1283},
abstract={This paper studies automatic image classification by modeling soft assignment in the popular codebook model. The codebook model describes an image as a bag of discrete visual words selected from a vocabulary, where the frequency distributions of visual words in an image allow classification. One inherent component of the codebook model is the assignment of discrete visual words to continuous image features. Despite the clear mismatch of this hard assignment with the nature of continuous features, the approach has been successfully applied for some years. In this paper, we investigate four types of soft assignment of visual words to image features. We demonstrate that explicitly modeling visual word assignment ambiguity improves classification performance compared to the hard assignment of the traditional codebook model. The traditional codebook model is compared against our method for five well-known data sets: 15 natural scenes, Caltech-101, Caltech-256, and Pascal VOC 2007/2008. We demonstrate that large codebook vocabulary sizes completely deteriorate the performance of the traditional model, whereas the proposed model performs consistently. Moreover, we show that our method profits in high-dimensional feature spaces and reaps higher benefits when increasing the number of image categories.},
keywords={feature extraction;image classification;automatic image classification;codebook model;image features;soft assignment modeling;visual word ambiguity;Computer vision;image/video retrieval.;object recognition},
doi={10.1109/TPAMI.2009.132},
ISSN={0162-8828},
month={July},}