Affiliation:
1. Justus-Liebig-Universität Gießen, Germany
2. NTNU-Norwegian University of Science and Technology, Norway
3. University of York, United Kingdom and NTNU-Norwegian University of Science and Technology, Norway
Abstract
Previous research investigated the perceptual dimensionality of achromatic reflection of opaque surfaces, by using either simple analytic models of reflection or measured reflection properties of a limited sample of materials. Here, we aim to extend this work to a broader range of simulated materials. In a first experiment, we used sparse multidimensional scaling techniques to represent a set of rendered stimuli in a perceptual space that is consistent with participants’ similarity judgments. Participants were presented with one reference object and four comparisons, rendered with different material properties. They were asked to rank the comparisons according to their similarity to the reference, resulting in an efficient collection of a large number of similarity judgments. To interpret the space individuated by multidimensional scaling, we ran a second experiment in which observers were asked to rate our experimental stimuli according to a list of 30 adjectives referring to their surface reflectance properties. Our results suggest that perception of achromatic reflection is based on at least three dimensions, which we labelled “Lightness,” “Gloss,” and “Metallicity,” in accordance with the rating results. These dimensions are characterized by a relatively simple relationship with the parameters of the physically based rendering model used to generate our stimuli, indicating that they correspond to different physical properties of the rendered materials. Specifically, “Lightness” relates to diffuse reflections, “Gloss” to the presence of high contrast sharp specular highlights, and “Metallicity” to spread out specular reflections.
Funder
Deutsche Forschungsgemeinschaft
Norges Forskningsråd
Publisher
Association for Computing Machinery (ACM)
Subject
Experimental and Cognitive Psychology,General Computer Science,Theoretical Computer Science
Cited by
22 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献