Range Image Segmentation Using a LEGION Network

This paper presents a neural network approach for range image segmentation. We use a locally excitatory globally inhibitory oscillator network (LEGION) as the framework for range image segmentation. Each oscillator in the LEGION network has excitatory lateral connections to the oscillators in its neighborhood as well as a connection with a global inhibitor. The feature detector associated with each oscillator estimates the surface normal and curvature at its corresponding pixel location. The lateral connection between two oscillators is established based on a similarity measure of their feature vectors. The emergent behavior of the LEGION network gives rise to the segmentation result. Unlike other methods, our scheme needs no assumption about the underlying structures in image data and no prior knowledge regarding the number of regions. Experimental results using real range images are presented.