Skip to main content
Han Wang

    Han Wang

    This paper describes the implementation of a 3D vision algorithm, Droid, on the Oxford parallel vision architecture, PARADOX, and the results of experiments to gauge the algorithm's effectiveness in providing navigation data for an... more
    This paper describes the implementation of a 3D vision algorithm, Droid, on the Oxford parallel vision architecture, PARADOX, and the results of experiments to gauge the algorithm's effectiveness in providing navigation data for an autonomous guided vehicle. The algorithm reconstructs 3D structure by analysing image sequences obtained from a moving camera. In this application, the architecture delivers a performance of greater than 1 frame per second — 17 times the performance of a Sun-4 alone.
    Localized feature points, particularly corners, can be computed rapidly and reliably in images, and they are stable over image sequences. Corner points provide more constraint than edge points, and this additional constraint can be... more
    Localized feature points, particularly corners, can be computed rapidly and reliably in images, and they are stable over image sequences. Corner points provide more constraint than edge points, and this additional constraint can be propagated effectively from corners along edges. Implemented algorithms are described to compute optic flow and to determine scene structure for a mobile robot using stereo or structure from motion. It is argued that a mobile robot may not need to compute depth explicitly in order to navigate effectively.