Statistics Explains Geometrical Optical Illusions

Azriel Rosenfeld has been our mentor for the last decade. In everyday conversation he stressed the importance of intuition, the unreasonable effectiveness of mathematics in understanding the world, and the power of simplicity inherent in deep ideas. Because we mostly worked on 3D vision, he often argued that 2D vision has been, is, and will continue to be a large source of problems. Strangely enough, we arrived at this study, which is our first on 2D vision, through problems we encountered in our work in 3D motion and shape. Azriel Rosenfeld was also one of the first to apply statistics [27, 28] to image analysis and understanding, and he always reminded us of the uncertainties involved in visual computations. This paper shows that statistics cannot be ignored, not even in the interpretation of two simple straight intersecting lines. It demonstrates that uncertainty in the visual data causes problems for the early visual processes. Because of noise, the estimation of features, such as lines, intersections of lines, and local image motion, is biased. The inevitability of this bias provides an explanation for many well-known geometrical optical illusions.

[1]  F. Campbell,et al.  Neurophysiological Localization of the Vertical and Horizontal Visual Coordinates in Man , 1970, Science.

[2]  E. Marg.  A VISION OF THE BRAIN , 1994 .

[3]  D Marr,et al.  Theory of edge detection , 1979, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[4]  L. Ganz.  Mechanism of the figural aftereffects. , 1966, Psychological review.

[5]  Tony Lindeberg,et al.  Scale-Space Theory in Computer Vision , 1993, Lecture Notes in Computer Science.

[6]  Chung Chiang,et al.  A new theory to explain geometrical illusions produced by crossing lines , 1968 .

[7]  Reinhard Tausch,et al.  Optische Täuschungen als artifizielle Effekte der Gestaltungsprozesse von Größen- und Formenkonstanz in der natürlichen Raumwahrnehmung , 1954 .

[8]  Hermann von Helmholtz,et al.  Treatise on Physiological Optics , 1962 .

[9]  J. Feldman.  Four frames suffice: A provisional model of vision and space , 1985, Behavioral and Brain Sciences.

[10]  Jean Piaget.  Les méchanismes perceptifs. , 1961 .

[11]  A. Kundt,et al.  Untersuchungen über Augenmaass und optische Täuschungen , 1863 .

[12]  R. W. Rodieck,et al.  Analysis of receptive fields of cat retinal ganglion cells. , 1965, Journal of neurophysiology.

[13]  A. H. Pierce.  Die verschobene Schachbrettfigur. , 2022 .

[14]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  V. Virsu.  Underestimation of curvature and task dependence in visual perception of form , 1971 .

[16]  E. Swift.  A New Visual Illusion of Direction. , 2022 .

[17]  H. Wallach,et al.  Figural aftereffects; an investigation of visual processes. , 1944 .

[18]  Yiannis Aloimonos,et al.  The Statistics of Optical Flow , 2001, Comput. Vis. Image Underst..

[19]  R. Gregory,et al.  Distortion of Visual Space as Inappropriate Constancy Scaling , 1963, Nature.

[20]  Y. J. Tejwani,et al.  Robot vision , 1989, IEEE International Symposium on Circuits and Systems,.

[21]  D. Hubel,et al.  Integrative action in the cat's lateral geniculate body , 1961, The Journal of physiology.

[22]  E L Thomas,et al.  Movements of the eye. , 1968, Scientific American.

[23]  J. O. Robinson.  The Psychology of Visual Illusion , 1972 .

[24]  A. Rosenfeld.  Geometrical probability in automatic pattern recognition , 1964 .

[25]  H. Carr.  An introduction to space perception , 1935 .

[26]  J. P. Southall.  Helmholtz's Treatise on Physiological Optics, Translated from the Third German Edition. Vol. I , 1925 .

[27]  V. Virsu.  Tendencies to eye movement, and misperception of curvature, direction, and length , 1971 .