This paper presents a new method for extracting pedestrian contours from images using 2D and 3D information obtained from a stereo-vision acquisition system. Two pedestrian contour types are extracted. First is obtained from static pedestrian confidence images using fixed background scenes and second from general traffic scenes having variable background. A robust approach for building contour ...

The evaluation of vision systems is not possible without parameter control as incorrect parameter adjustment can cause performance losses that are more signiicant than the eeect of the component being tested. In this paper we present a model for the contour extraction process and derive from it a method for automatic parameter control. The method works with arbitrary edge detectors, edge types ...

This paper presents a three-module strategy for edge detection. The first and the last module involve well-known methods: the first module is a parallel process computing local edge strength and direction while the last module is a sequential process following edges. The originality of the overall method resides in the intermediate module, seen as a generalization of the nonmaximum deletion alg...