Abstract
Shape recovery from a monocular image with errors is addressed. Shape recovery necessitates the use of additional plausible constraints on typical structures and features of the objects in an ordinary scene. We propose an hypothesization and verification method for 3D shape recovery based on geometrical constraints peculiar to man-made objects. 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. Another lies in dealing with the error by preliminary processes to extract 2D geometrical features from a real image. A concurrent mechanism has been implemented which is based on energy minimization using a parallel network for relaxation, and is capable of maintaining consistency between constraint assignment and shape recovery. The error in an input image is also corrected through the process of shape recovery.
Preview
Unable to display preview. Download preview PDF.
References
J.J.Hopfield and D.W.Tank: “Neural” Computation of Decisions in Optimization Problems. Biological Cybernetics 52, 141–152 (1985)
R.Horaud: New Methods for Matching 3-D Objects with Single Perspective Views. IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI-9, 401–412 (1987)
Y.Shirai: Three-Dimensional Computer Vision. Berlin: Springer 1987
B.Irie, S.Miyake: Capabilities of Three-layered Perceptrons. Proceedings of IEEE Annual International Conference on Neural Networks 1, 641–648 (1988)
H.Kawahara, T.Irino: A Procedure for Designing 3-Layer Neural Networks Which Approximate Arbitrary Continuous Mapping: Applications to Pattern Processing. IEICE Techical Report MBE88-54, 47–54 (1989) (in Japanese)
S.Kirkpatrik, C.D.Gelatt and Jr.M.P.Vecchi: Optimization by Simulated Annealing. Science 220, 671–680 (1983)
F.Ulupinar and R.Nevatia: Constraints for Interpretation of Line Drawings under Perspective Projection. CVGIP Image Understanding 53, 88–96 (1991)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1992 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kakusho, K., Dan, S., Abe, N., Kitahashi, T. (1992). Shape recovery and error correction based on hypothetical constraints by parallel network for energy minimization. In: Nakamura, A., Nivat, M., Saoudi, A., Wang, P.S.P., Inoue, K. (eds) Parallel Image Analysis. ICPIA 1992. Lecture Notes in Computer Science, vol 654. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-56346-6_37
Download citation
DOI: https://doi.org/10.1007/3-540-56346-6_37
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-56346-4
Online ISBN: 978-3-540-47538-5
eBook Packages: Springer Book Archive