Appearance Based Object Recognition with a Large Dataset using Decision Trees
Patricia Blackwell, David E. Austin · 2004
A framework for a new object recognition system is presented that tailors the classic decision tree to the specific needs of object recognition. The key advantage of such a system is the ability to handle a large database of known objects with real-time performance, a failing of current object recognition systems. A new image database is presented that is challenging both in size (over 100,000 images), and content (highly similar objects). Early results, using the simplest features, are promising with a recognition rate of 89%. 1