밀결합 방법을 이용한 스테레오 변위와 광류의 융합

Abstract

As a convergence method for stereo matching and motion estimation, this paper presents an equation, called Disparity-Optical flow Equation (DOE), that relates disparity with optical flow in recti&#64257

ed images. Considering this equation as a constraint, this paper suggests two algorithms: first algorithm uses a recursive filter in which the two vision quantities are determined alternatively by time and measurement updates

the second algorithm determines the disparity and optical flow simultaneously by full rank constraint equations. Both algorithms determines the disparity and optical flowow efficiently. These algorithms are completely different from the previous approaches that try to compromise the outputs of the two modules.The experiments show that the algorithms can resolve the ambiguous situations in which either stereo matching or motion estimation fail to give satisfactory results.