Paper Info
Title
Latent Dirichlet Allocation Models for Image Classification
Abstract
Two new extensions of latent Dirichlet allocation (LDA), denoted topic-supervised LDA (ts-LDA) and class-specific-simplex LDA (css-LDA), are proposed for image classification. An analysis of the supervised LDA models currently used for this task shows that the impact of class information on the topics discovered by these models is very weak in general. This implies that the discovered topics are driven by general image regularities, rather than the semantic regularities of interest for classification. To address this, ts--LDA models are introduced which replace the automated topic discovery of LDA with specified topics, identical to the classes of interest for classification. While this results in improvements in classification accuracy over existing LDA models, it compromises the ability of LDA to discover unanticipated structure of interest. This limitation is addressed by the introduction of css-LDA, an LDA model with class supervision at the level of image features. In css-LDA topics are discovered per class, i.e., a single set of topics shared across classes is replaced by multiple class-specific topic sets. The css-LDA model is shown to combine the labeling strength of topic-supervision with the flexibility of topic-discovery. Its effectiveness is demonstrated through an extensive experimental evaluation, involving multiple benchmark datasets, where it is shown to outperform existing LDA-based image classification approaches.
Year
DOI
Venue
2013
10.1109/TPAMI.2013.69
Pattern Analysis and Machine Intelligence, IEEE Transactions
Keywords
Field
DocType
category theory,image classification,solid modelling,LDA-based image classification approaches,automated topic discovery,class information,class supervision,class-specific topic sets,class-specific-simplex LDA,classification accuracy,css-LDA model,css-LDA topics,experimental evaluation,image features,image regularity,labeling strength,latent Dirichlet allocation models,multiple benchmark datasets,semantic regularity,supervised LDA models,topic-discovery,topic-supervised LDA,topic-supervision,ts-LDA,Image classification,attributes,graphical models,latent Dirichlet allocation,semantic classification
Latent Dirichlet allocation,Pattern recognition,Computer science,Artificial intelligence,Graphical model,Contextual image classification
Journal
Volume
Issue
ISSN
35
11
0162-8828
Citations 
PageRank 
References 
30
0.97
14
Authors
2
Name
Order
Citations
PageRank
N Rasiwasia1117334.61
Nuno Vasconcelos25410273.99