Abstract
A statistical correlation technique (SCT) and two variants of a neural network are presented to solve the motion correspondence problem. Solutions of the motion correspondence problem aim to maintain the identities of individuated elements as they move. In a pre-processing stage, two snapshots of a moving scene are convoluted with two-dimensional Gabor functions, which yields orientations and spatial frequencies of the snapshots at every position. In this paper these properties are used to extract, respectively, the attributes orientation, size and position of line segments. The SCT uses cross-correlations to find the correct translation components, angle of rotation and scaling factor. These parameters are then used in combination with the positions of the line segments to calculate the centre of motion. When all of these parameters are known, the new positions of the line segments from the first snapshot can be calculated and compared to the features in the second snapshot. This yields the solution of the motion correspondence problem. Since the SCT is an indirect way of solving the problem, the principles of the technique are implemented in interactive activation and competition neural networks. With boundary problems and noise these networks perform better than the SCT. They also have the advantage that at every stage of the calculations the best candidates for corresponding pairs of line segments are known.
Similar content being viewed by others
References
Braddick OJ (1974) A short-range process in apparent motion. Vision Res 14:519–527
Braddick OJ (1980) Low-level and high-level processes in apparent motion. Philos Trans R Soc Lond [Biol] 290:137–151
Canny J (1986) A computational approach to edge detection. IEEE Trans Pattern Analysis Machine Intell 8:679–698
Cavanagh P, Mather G (1989) Motion: the long and short of it. Spatial Vision 4:103–129
Dawson MRW (1991) The how and why of what went where in apparent motion: modelling solutions to the motion correspondence problem. Psych Rev. 98:569–603
Dinic EA, Kronrod MA (1969) An algorithm for the solution of the assignment problem. Sov Math Dokl 10 (6):1324–1326
Grzywacz NM, Yuille AL (1988) Massively parallel implementations of theories for apparent motion. Spatial Vision 3:15–44
Heitger F, Rosenthaler L, Von der Heydt R, Peterhans E, Kübler O (1992) Simulation of neural contour mechanisms: from simple to end-stopped cells. Vison Res 32:963–981
Hopfield JJ, Tank DW (1985) “Neural” computation of decisions in optimization problems. Biol Cybern 52:141–152
Marr D (1976) Early processing of visual information. Philos Trans R Soc Lond [Biol] 275:483–519
Marr D, Hildreth E (1980) Theory of edge detection. Proc R Soc Lond [B] 207:187–217
McClelland, Rumelhart (1988) Explorations in parallel distributed processing. A handbook of models, programs and exercises. MIT Press, Cambridge
Petersik JT (1989) The two-process distinction in apparent motion. Psych. Bulletin 106:107–127
Schuling FH, Altena P, Mastebroek HAK (1990) The computational measurement of apparent motion: a recurrent pattern recognition strategy as an approach to solve the correspondence problem. Biol Cybern 62:463–473
Spreeuwers LJ (1991) A neural network edge detector. Proceedings, SPIE/SPSE Symposium on Electrical Imaging Science & Techology, San Jose, Calif, 1451–16
Spreeuwers LJ (1992) Image filtering with neural networks. PhD Thesis. University of Twente
Ullman S (1979) The interpretation of visual motion. MIT Press, Cambridge
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
van Deemter, J.H., Mastebroek, H.A.K. A statistical correlation technique and a neural network for the motion correspondence problem. Biol. Cybern. 70, 329–344 (1994). https://doi.org/10.1007/BF00200330
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF00200330