Shape from Multiple Cues: Integrating Local Brightness Information

In this paper, a novel approach to shape from multiple cues is further refined and applied to the task of combining contours with local brightness information. The approach, furtheron called the sculptor's principle, does not attempt to construct an 'optimum' solution but rather finds an enclosing volume that must contain the true solution. Shape from contours algorithms only provide accurate results for points on the convex hull of object. The shape ambiguity at concave parts, however, has to be removed by integrating complementary shape cues. This is addressed by a new idea of how to exploit the local brightness information, the luminance, for reconstruction purposes. The shape from luminance algorithm is illustrated on the enclosing volume of a dummy.

[1]  L. Quam Hierarchical warp stereo , 1987 .

[2]  Thomas Fromherz,et al.  Shape from contours as initial step in shape from multiple cues , 1994, Other Conferences.

[3]  Tomoharu Nakahara,et al.  A Multiple-baseline Stereo Method , 2002 .

[4]  H.A. Beyer,et al.  Accurate calibration of CCD-cameras , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  V. Leitáo,et al.  Computer Graphics: Principles and Practice , 1995 .

[6]  Guna S. Seetharaman Image sequence analysis for three-dimensional perception of dynamic scenes , 1994 .