Hilfe Wegweiser Datenschutzhinweis Impressum Kontakt





Surface-slant and surface-curvature from texture.


Rosas,  P
Max Planck Institute for Biological Cybernetics, Max Planck Society;

Schepers J, Wichmann,  FA
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Externe Ressourcen
Es sind keine Externen Ressourcen verfügbar
Volltexte (frei zugänglich)
Es sind keine frei zugänglichen Volltexte verfügbar
Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar

Rosas, P., Schepers J, Wichmann, F., & Wagemans, J. (2002). Surface-slant and surface-curvature from texture. Poster presented at 25th European Conference on Visual Perception, Glasgow, UK.

The contribution of texture to the perception of slant and curvature was studied in two experiments in which we collected judgments of human subjects by means of probe adjustment. For the slant-perception experiment, our stimuli consisted of flat surfaces receding away from the subject, while for the curvature experiment we used surfaces depicting half-elliptic cylinders whose major semi-axes pointed towards the subject. Both types of stimuli were generated under perspective projection with purpose written software. The surfaces were mapped with different types of textures: uniform dot lattices, polka dots, Voronoi tessellations, orthogonal sinusoidal plaid patterns, fractal or 1/f noise, 'coherent' noise, and a 'diffusion-based' texture (leopard-skin-like). Although we observed differences in performance for different textures, the results generally show a higher correlation between the perceived and depicted slant of a plane than between perceived and depicted curvature. These results suggest that texture may be more informative for perceiving slant than curvature. For the plaid textures, however, qualitatively similar results were obtained in both experiments. Since this is not the case for textures comparatively similar in uniformity (eg dot lattice), further spatial-frequency-based comparisons might show relevant differences for a spectrum-based shape-from-texture mechanism.