An extension of the latent Dirichlet allocation (LDA), denoted class-specific-simplex LDA (css-LDA), is 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, we introduce a model that induces supervision in topic discovery, while retaining the original flexibility of LDA to account for unanticipated structures of interest. The proposed css-LDA, is 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 proposed model can be used for generative classification using Bayes rule or even extended to discriminative classification. css-LDA model is shown to endow a vector of class and topic specific count statistics similar to the Bag-of-words (BoW). This image representation could then be used in discriminant learning. The effectiveness of css-LDA model in both generative and discriminative classification frameworks is demonstrated through an extensive experimental evaluation, involving multiple benchmark datasets, where it is shown to outperform existing LDA and BoW based image classification approaches.