In this paper, we present an object recognition approach that in addition allows to discover intra-class modalities exhibiting highcorrelated visual information. Unlike to more conventional approaches based on computing multiple specialized classifiers, the proposed approach combines a single classifier, Boosted Random Ferns (BRFs), with probabilistic Latent Semantic Analysis (pLSA) in order to recognize an object class and to find automatically the most prominent intra-class appearance modalities (clusters) through tree-structured visual words. The proposed approach has been validated in synthetic and real experiments where we show that the method is able to recognize objects with multiple appearances.


computer vision, object detection, pattern clustering.

Author keywords

boosting, random ferns, tracking, clustering, plsa, multimodal

Scientific reference

M. Villamizar, A. Garrell Zulueta, A. Sanfeliu and F. Moreno-Noguer. Multimodal object recognition using random clustering trees, 7th Iberian Conference on Pattern Recognition and Image Analysis, 2015, Santiago de Compostela, in Pattern Recognition and Image Analysis, Vol 9117 of Lecture Notes in Computer Science, pp. 496-504, 2015, Springer.