Parallel algorithm for depth perception from radial optical flow fields

Abstract

While optical flow has been often proposed for guiding a moving robot, its computational complexity has mostly prevented its actual use in real applications. We describe a restricted form of optical flow algorithm, which can be parallelized on chain-like neuronal structures, combining simplicity and speed. In addition, this algorithm makes use of predicted motion trajectories in order to remove noise from the input images.