Interactively Training Pixel Classiiers

| For typical classiication tasks, all training data are prepared in advance and are supplied to the classiier all at once. This is unnecessarily expensive and incurs overrtting problems, since the individual contributions of the training instances to the classiier are not known. We address this by proposing an interactive incremental framework for image classiier construction , where small numbers of training examples are supplied at each user interaction. After incorporating new training instances, the classiier immediately reclassiies the image to provide the user with instant feedback. This allows the user to choose additional informative training pixels from among the currently misclassiied ones. Using a realistic terrain classiica-tion task, we demonstrate the potential of our method to generate small and accurate decision tree classiiers from surprisingly few training examples while avoiding overspecialization. We also briefly discuss the novel concept of hierarchical classiication, where higher-level classiiers take as input the output of lower-level classiiers. We present preliminary results indicating that within our interactive framework, this is a practical approach to exploiting spatial relationships for classiication.