Motion segmentation based on motion/brightness integration and oscillatorycorrelation

A segmentation method based on the integration of motion and brightness is proposed for image sequences. The method is composed of two parallel pathwa ys that process motion and brightness, respectively. Inspired by the visual system, the motion pathway has two stages. The first stage estimates local motion at locations with reliable information. The second stage performs s egmentation based on local motion estimates. In the brightness pathway, the input scene is segmented into regions based on brightness distribution. Su bsequently, segmentation results from the two pathways are integrated to re fine motion estimates. The final segmentation is performed in the motion ne twork based on refined estimates. For segmentation, locally excitatory glob ally inhibitory oscillator network (LEGION) architecture is employed whereb y the oscillators corresponding to a region of similar motion/brightness os cillate in synchrony and different regions attain different phases. Results on synthetic and real image sequences are provided, and comparisons with o ther methods are made.