Abstract
We studied whether human observers can estimate the illumination direction from 3D textures of random Brownian surfaces, containing undulations over a range of scales. The locally Lambertian surfaces were illuminated with a collimated beam from random directions. The surfaces had a uniform albedo and thus texture appeared only through shading and shadowing. The data confirm earlier results with Gaussian surfaces, containing undulations of a single scale. Observers were able to accurately estimate the source azimuth. If shading dominated the images, the observers committed 180° errors. If cast shadows were present, they resolved this convex-concave-ambiguity almost completely. Thus, observers relied on second-order statistics in the shading regime and used an unidentified first-order cue in the shadow regime. The source elevations could also be estimated, which can be explained by the observers’ exploitation of the statistical homogeneity of the stimulus set. The fraction of the surface that is in shadow and the median intensity are likely cues for these elevation estimates.
Highlights
IntroductionLuminance distribution or histogram-based cues are, for instance, the width, average, and skewness of the luminance distribution (Ho, Landy, & Maloney, 2008; Kim & Anderson, 2010; Motoyoshi, Nishida, Sharan, & Adelson, 2007; Pont & Koenderink, 2005, 2008; Wijntjes & Pont, 2010)
We will consider texture due to the illumination of rough surfaces
In another study (Koenderink et al, 2003), we tested whether human observers were able to carry out this task for frontally viewed real 3D textures from the Curet database, see, for examples, Figure 1 left, and we found that the experimental results were surprisingly similar to the theoretical predictions
Summary
Luminance distribution or histogram-based cues are, for instance, the width, average, and skewness of the luminance distribution (Ho, Landy, & Maloney, 2008; Kim & Anderson, 2010; Motoyoshi, Nishida, Sharan, & Adelson, 2007; Pont & Koenderink, 2005, 2008; Wijntjes & Pont, 2010). For arbitrary textures (i.e., statistically inhomogeneous sets of textures), the elevation cannot be estimated due to the bas-relief ambiguity (Belhumeur, Kriegman, & Yuille, 1997)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
