The Shape of Illusory Figures

We have been developing a stochastic model for figure-ground separation[9][3][12]. The model selects/constructs theforeground with preferenceforfigures with "moreconvex" shapes. When these models are applied to illusory figures ([7]) they yield perceptually accurate selection offigure and background. The approach is based on an "entropy" measure of a region dijfksion Markov modelfrom a set of localjigure/ground hypothesis. The contour boundaries are implicitly represented,via the thresholding of the difSusionresult. What optimalproperties do the illusory contours satisfies ? We show that the entropy criteria selects contours such as to minimize a Taylor series of the even derivatives with respect to the length of the contour: The coefJicients are positive and they get exponentially smaller as the derivatives increase. The zeroth order term suggest that small length contours arepreferred, the second order terms suggests that curvature-like term is minimized (with le...

• Type: Conference
• Year: 1999
• Where: ICIP
• Authors: Davi Geiger, Krishnan Kumaran, Hsing-Kuo Pao, Nava Rubin
• Keywords: Criteria Selects Contours, ICIP 1999, Illusory Contours Satisfies, Image Processing, Order Terms