Purchase this article with an account.
Huseyin Boyaci, Laurence T. Maloney; The effect of an illuminant direction cue based on cast shadows on lightness perception in three-dimensional scenes. Journal of Vision 2004;4(8):121. doi: 10.1167/4.8.121.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The intensity of light emitted by a matte surface depends on its orientation with respect to punctate light sources in the scene. We recently reported that observers discount orientation in making judgments of surface albedo (‘lightness’) in binocularly viewed scenes illuminated by punctate and diffuse light sources (Boyaci et al, 2003, JOV, 3, 541–553). To do so, they must have access to an accurate estimate of the direction to the punctate source. There are several cues that could provide this information: cast shadows, attached shadows and specular highlights. We examined whether a cue based on cast shadows affects observers judgments of albedo. The stimuli were stereo image pairs of scenes presented in a Wheatstone stereoscope. Each pair was computer rendered with a diffuse-punctate lighting model. All scenes contained an achromatic matte test patch. The orientation of the test patch varied across trials: −60,−45,0,45,60 deg (0 was fronto-parallel). We developed a method for rendering scenes so that cast shadow orientations signaled azimuth angles to the punctate source that were perturbed by −20, −10,0,10,20, and 30 deg (0 was unperturbed) while other cues still signaled the true value. Observers matched the apparent albedo of the test patch to a standard gray scale. The total number of trials for each of four observers was 600: 6×5×20. We found that in the unperturbed scenes all observers discounted orientation (p<.005). Two observers showed no change in performance in response to perturbations, indicating that they were not making use of the cast shadow cue. The other two reacted to non-zero perturbations by ceasing to discount orientation in estimating albedo. While it is generally accepted that human observers are insensitive to inconsistencies in cast shadows, the mechanisms underlying albedo estimation for these two observers appear to be exquisitely sensitive to inconsistencies between cast shadow cues and other cues to illuminant direction.
Grant EY08266 from the National Institute of Health; Grant RG0109/1999-B from the Human Frontiers Science Program
This PDF is available to Subscribers Only