Finding continuous-model moving shapes by evidence gathering

Two recent approaches are combined in a new technique to find moving arbitrary shapes. We combine the new velocity Hough transform, which extracts moving conic sections, with a continuous formulation for arbitrary shape extraction. The new velocity Hough transform is demonstrated to provide motion analysis that is resilient to noise and to be able to detect its target shapes, which are both moving and arbitrary. Further, it is shown to have performance advantage over contemporaneous single-image extraction techniques. Finally, it appears to offer improved immunity to noise and occlusion, as consistent with evidence gathering techniques. The target application of this new approach is for use in automatic gait recognition, as a primer for shape extraction or as a basis for a recognition technique.