Abstract
Shape recovery from a monocular image is addressed. It is often said that the information conveyed by an image is insufficient to reconstruct 3D shapes of objects in the image. This implies that shape recovery from an image necessitates the use of additional plausible constraints on typical structures and features of the objects in an ordinary scene. We propose a hypothesization and verification method for 3D shape recovery based on geometrical constraints peculiar to man-made objects. The objective is to increase the robustness of computer vision systems. One difficulty with this method lies in the mutual dependency between proper assignment of constraints to the regions in a given image and recovery of a consistent 3D shape. A concurrent mechanism has been implemented which is based on energy minimization using a parallel network for relaxation. This mechanism is capable of maintaining consistency between constraint assignment and shape recovery.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
M.Ishizuka, Pattern cognition and hypothesis-based reasoning, IEICE Tech. Reprt. PRU 87–123: 33–40, 1988 (in Japanese).
Y.Watanabe, N.Abe, and T.Kitahashi, Three-dimensional interpretation of image regions using assumptions with preference, IEICE Tech. Reprt. IE 88–127: 9–16, 1989 (in Japanese).
R.Horaud, New methods for matching 3-D objects with single perspective views, IEEE Trans. Patt. Anal. Mach. Intell. 9(3): 401–412, 1987.
Y.Shirai, Three-Dimensional Computer Vision, Springer-Verlag: Berlin, Heidelberg, 1987.
T.Tanaka, T.Kawashima, and K.Kanatani, 3D recovery of polyhedra by parallelism heuristics, Trans. IEICE J72-D II(4): 517–525, 1989 (in Japanese).
K.Maehara, T.Kawashima, and K.Kanatani, 3D recovery of polyhedra by rectangularity heuristics, Trans. IEICE J72-D II(6): 887–895, 1989 (in Japanese).
J.J.Hopfield and D.W.Tank, “Neural” computation of decisions in optimization problems, Biolog. Cybern. 52: 141–152, 1985.
K.Sakaue and N.Yokoya, Relaxation and regularization, J. Inf. Process. 30(9): 1047–1057, 1989 (in Japanese).
K. Sugihara, Algebraic approach to the recovery of three-dimensional shapes from single images, Trans. IECE J66-D(5): 541–548, 1983 (in Japanese).
T.Poggio, V.Torre, and C.Koch, Computational vision and regularization theory, Nature 317(6035): 314–319, 1985.
B.Irie and S.Miyake, Capabilities of three-layered perceptrons, Proc. Annu. Intern. Conf. Neural Networks 1: 641–648, San Diego, 1988.
K.Funahashi, On the capabilities of neural networks, IEICE Tech. Reprt. MBE88 52: 127–134, 1988 (in Japanese).
H.Kawahara, T.Irino, A procedure for designing 3-layer neural networks which approximate arbitrary continuous mapping: Applications to pattern processing, IEICE Tech. Reprt. MBE88 54: 47–54, 1989 (in Japanese).
S.Geman and D.Geman, Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images, IEEE Trans. Patt. Anal. Mach. Intell., 6(6): 721–741, 1984.
C.Koch, J.Marroquin, and A.Yuille, Analog “neuronal” networks in early vision, Proc. Natl. Acad. Sci. USA, 83: 4263–4267, 1986.
S.Kirkpatrik, C.D.Gelatt, and M.P.VecchiJr., Optimization by simulated annealing, Science, 220(4598): 671–680, 1983.
M.Kawato, T.Ikeda, and S.Miyake, Learning in neural networks for visual information processing, J. ITEJ, 42(9): 918–924, 1988 (in Japanese).
F.Ulupinar and R.Nevatia, Constraints for interpretation of line drawings under perspective projection, CVGIP Image Understanding 53(1): 88–96, 1991.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Kakusho, K., Dan, S., Kitahashi, T. et al. Computer vision based on a hypothesization and verification scheme by parallel relaxation. Int J Comput Vision 9, 13–30 (1992). https://doi.org/10.1007/BF00163581
Issue Date:
DOI: https://doi.org/10.1007/BF00163581