Invariant surface segmentation through energy minimization with discontinuities
Tóm tắt
The computational problems in segmenting range data into surface patches based on the invariant surface properties, i.e., mean curvature H and Gaussian curvature K, are investigated. The goal is to obtain reliable HK surface maps. Two commonly encountered problems are: firstly the noise effect in computing derivative estimates, and secondly the smoothing across discontinuities. Here, the segmentation is formulated as finding minimization solutions of energy functionals involving discontinuities. A two-stage approach to the goal is presented: stage (1) from a range image to curvature images and stage (2) from the curvature images to the HK maps. In both stages, solutions are found through minimizing energy functionals that measure the degree of bias of a solution from two constraints: the closeness of the solution to the data, and the smoothness of the solution controlled by predetermined discontinuities. Propagation across discontinuities is prevented during minimization, which preserves the original surface shapes. Experimental results are given for a variety of test images.
Tài liệu tham khảo
Besl, P.J., 1988. Surfaces in Range Image Understanding, Springer-Verlag: New York.
Besl, P.J. and Jain, R.C., 1986. Invariant surface characteristics for three dimensional object recognition in range images. Comput. Vision Graphics and Image Process, 33 (1):33–80.
Blake, A., and Zisserman, A. 1987. Visual Reconstruction, MIT press: Cambridge, MA.
Brady, M., Ponce, J., Yuille, A. and Asada, H. 1985. Describing surfaces. Comput. Vision Graphics and Image Process. 32:1–28.
Courant, R. and Hilbert, D., 1953, Method of Mathematical Physics, vol. #1. Interscience: New York.
Fan, T.J., Medioni, G., and Nevatia, R. 1987 Surface segmentation and description from curvature features. In Proc. DARPA Image Understanding Workshop, Los Angeles, February 23–25. pp. 351–359.
Paugeras, O.D., and Hebert, M., 1986. The representation recognition and locating of 3-D objects. Intern. J. Robotics Res. 5(3):27–52.
Geman, G., and Geman, D., 1984. Stochastic relaxation, Gibbs distribution and Bayesian restoration of images, IEEE Trans. Patt. Analy. Mach. Intell, PAMI6:721–741.
Grimson, W.E.L., 1984. From Images to Surfaces: A Computational Study of the Human Early Visual System. MIT Press: Cambridge, MA.
Haralick, R.M., Watson, L.T., and Laffey, T.J., 1983. The topographic primal sketch. Intern. J. Robotics Res., 2(1):50–72.
Hildreth, E.C., 1983. The Measurement of Visual Motion. MIT Press: Cambridge, MA.
Hoffman, R., and Jain, A.K. 1987. Segmentation and classification of range images. IEEE Trans. Patt. Analy. Mach. Intell. PAMI-10 (5):608–620.
Kirkpatrick, S., Gellatt, C.D., and Vecchi, M.P. 1983, Optimization by simulated annealing. Science, 220:671–680.
Koch, C., Marroquin, J., and Yuille, A., 1985. Analog ‘neuronal’ networks in early vision. A.I. Lab. Memo, No. 751, MIT, Cambridge, MA.
Leclerc, Y.G., and Zucker, S.W. 1987. The local structure of image discontinuities in one dimension. IEEE Trans. Patt. Anal. Mach. Intell. PAMI-9 (3).
Lee, D., and Pavlidis, T. 1987. One dimensional regularization with discontinuities. In Proc. Ist Intern. Conf. on Comput. Vision, London, 572–577.
Li, S.Z. 1989. A curve analysis approach to surface feature extractin from range image. In Proc. Intern. Workshop on Mach. Intell. and Vision, Tokyo, April.
Li, S.Z. 1989. 3D object recognition from range images: Computational framework and neural networks. In Proc. 2nd Intern. Conf. on Intelligent Autonomous Systems. Amsterdam, Netherlands, December 11–14, pp. 572–577.
Li, S.Z. 1990. Inexact matching of surfaces. VSSP-TR-3/90, E.E. Dept., Surrey Univ., February.
Lipschutz, M., 1969. Differential Geometry. McGraw-Hill: New York.
Liu, S.C., and Harris, J.G. 1989. Generalized smoothing networks in early vision. In Proc. CVPR'89, pp. 184–191.
Marroquin, J.L. 1985. Probabilistic solution of inverse problems. A.I. Tech. Report, No. 860, MIT, Cambridge, MA.
Medioni, G., and Nevatia, R. 1984. Description of 3-D surfaces using curvature properties. In Proc. DARPA Image Understanding Workshop, NewOrleans, LA October 3–4, pp. 291–299.
Poggio, R. and Koch, C. 1985. Ill-posed problems in early vision: from computational theory to analog networks. Proc. Roy. Soc. London, B 226, 303–323.
Poggio, T., Torre, V., and Koch, C. 1985. Computational vision and regularization theory. Nture, 317: 314–319.
Rosenfeld, A., Hummel, R., and Zucker, S. 1976. Scene labeling by relaxation operations. IEEE Trans. Syst., Man. Cybern, SMC-6:420.
Serra, J. 1982. Image Analysis and Mathematical Morphology. Academic Press: San Diego.
Smith, G.D. Numerical Solutions of Partial Differential Equations: Finite Difference Method. Oxford University Press: New York.
Terzopoulos, D. 1983. The role of constraints and discontinuities in visible surface reconstruction. In Proc. 8th Intern. Joint Conf. Artif. Intell. Karlsruhe, W. Germany, pp. 1073–1077.
Terzopoulos, d. 1986. Regularization of inverse problems involving discontinuities. IEEE Trans. Patt. Anal. Mach. Intell. PAMI-8:129–139.
Terzopoulos, D. 1988. Computation of visible-surface representation. IEEE Trans. Patt. Anal. Mach. Intell. PAMI-40:417–438.
Tikhnov, A.N. and Arsenin, V.A. 1977. Solutions of Ill-posed Problems. Winston & Sons: Washington, D.C.
Toore, V. and Poggio, T., 1984. On edge detection. A. I. Lab. Memo., No 768, MIT, Cambridge, MA, March.
Witkin, A., Terzopoulos, D., and Kass, M. 1987. Signal matching through scale space. Intern. J. Comput. Vision, 133–144.
Yuille, A. 1987. Energy function for early vision and analog networks. A.I. Lab. Memo, No. 987, MIT, Cambridge, MA, November.