Abstract. This paper proposes a new approach to learning a discriminative model of object classes, incorporating appearance, shape and context information efficiently. The learned ...
Jamie Shotton, John M. Winn, Carsten Rother, Anton...
For effective retrieval of visual information, statistical learning plays a pivotal role. Statistical learning in such a context faces at least two major mathematical challenges: ...
This paper presents a new algorithm for the automatic recognition of object classes from images (categorization). Compact and yet discriminative appearance-based object class mode...
Detecting shadows from images can significantly improve the performance of several vision tasks such as object detection and tracking. Recent approaches have mainly used illuminat...
Classification is one of the most essential tasks in data mining. Unlike other methods, associative classification tries to find all the frequent patterns existing in the input...