Abstract
Shading in images of 3-D objects refers to variations in luminance or color as a function of the surface orientation with respect to the illumination and viewing directions. A shaded image often gives rise to a compelling sense of 3-D shape, surface and illumination properties. Contrary to this intuitive sense that shading elicits, in theory, it is not yet certain how this is achieved. We used local attitude and cross-section adjustment procedures to examine whether the perception of shape from shading can be influenced by type of material surfaces and light field. Stimuli were renderings of 3-D convex, solid objects projected onto a CRT screen. Four different types of material surface reflectance (diffuse, specular, backscattering, asperity scattering BRDFs (bi-directional reflectance distribution function)), three different lighting modes (collimated, hemispherically diffuse and combination of collimated and Ganzfeld lighting), and four different illumination directions (0, 30, 60 and 90 degree elevations) were used to render the surface of the objects (sphere or ellipsoid). Observers' task was to adjust the attitude of a probe to match perceived attitude of 7 test locations on the image and then to set the extent of perceived protuberance of these points. Results from the two tasks showed that surfaces having the backscattering and asperity reflection tended to be perceived as flatter than those having the diffuse and specular scattering. However, the lighting mode and light directions had no effect on the perceived orientation of the shape. This suggests that 3-D surface relief is not affected by changes in the distribution of intensity caused by the extrinsic factors such as lighting modes and directions, but by those in the material surfaces. In conclusion, shape constancy does not hold for images rendered with different types of material surfaces.