Learning to recognize objects with little supervision
Peter Carbonetto, Gyorgy Dorko, Cordelia Schmid, Hendrik Kück and Nando de Freitas
International Journal of Computer Vision 2005.

Abstract

This paper shows (i) improvements over state of­the­art local feature recognition systems, (ii) how to formulate principled models for automatic local feature selection in object class recognition when there is lit­ tle supervised data, and (iii) how to formulate sensible spatial image context models using a conditional random field for integrating local features and segmentation cues (superpixels). By adopting sparse kernel methods, Bayesian learning techniques and data association with constraints, the proposed model identifies the most relevant sets of local features for recognizing object classes, achieves performance comparable to the fully supervised setting, and consistently outperforms existing methods for image classification.

[Edit]