SEEMORE: a view-based approach to 3-D object recognition using multiple visual cues | IEEE Conference Publication | IEEE Xplore