Analyzing appearance and contour based methods for object categorization

Abstract

Object recognition has reached a level where we can iden- tify a large number of previously seen and known objects. However, the more challenging and important task of cat- egorizing previously unseen objects remains largely un- solved. Traditionally, contour and shape based methods are regarded most adequate for handling the generaliza- tion requirements needed for this task. Appearance based methods, on the other hand, have been successful in object identification and detection scenarios. Today little work is done to systematically compare existing methods and char- acterize their relative capabilities for categorizing objects. In order to compare different methods we present a new database specifically tailored to the task of object catego- rization. It contains high-resolution color images of 80 ob- jects from 8 different categories, for a total of 3280 images. It is used to analyze the performance of several appearance and contour based methods. The best categorization re- sult is obtained by an appropriate combination of different methods.