Learning a discriminative classifier using shape context distances

For purpose of object recognition, we learn one discriminative classifier based on one prototype, using shape context distances as the feature vector. From multiple prototypes, the outputs of the classifiers are combined using the method called "error correcting output codes". The overall classifier is tested on benchmark dataset and is shown to outperform existing methods with far fewer prototypes.

• Type: Conference
• Year: 2003
• Where: CVPR
• Authors: Hao Zhang 0003, Jitendra Malik
• Keywords: Computer Vision, CVPR 2003, Error Correcting Output, Fewer Prototypes, Overall Classifier, Shape Context Distances