Probabilistic models for supervised dictionary learning

• Published in: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition pp. 2305 - 2312
• Main Authors: Xiao-Chen Lian, Zhiwei Li, Changhu Wang, Bao-Liang Lu, Lei Zhang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2010
• Subjects: Costs; Dictionaries; Europe; Image representation; Kernel; Large-scale systems; Machine learning; Principal component analysis; Support vector machine classification; Support vector machines
• ISBN: 1424469848; 9781424469840
• ISSN: 1063-6919; 1063-6919
• DOI: 10.1109/CVPR.2010.5539915

Dictionary generation is a core technique of the bag-of-visual-words (BOV) models when applied to image categorization. Most of previous approaches generate dictionaries by unsupervised clustering techniques, e.g. k-means. However, the features obtained by such kind of dictionaries may not be optimal for image classification. In this paper, we propose a probabilistic model for supervised dictionary learning (SDLM) which seamlessly combines an unsupervised model (a Gaussian Mixture Model) and a supervised model (a logistic regression model) in a probabilistic framework. In the model, image category information directly affects the generation of a dictionary. A dictionary obtained by this approach is a trade-off between minimization of distortions of clusters and maximization of discriminative power of image-wise representations, i.e. histogram representations of images. We further extend the model to incorporate spatial information during the dictionary learning process in a spatial pyramid matching like manner. We extensively evaluated the two models on various benchmark dataset and obtained promising results.