Perception of 3D shape and material properties from patterns of image shading

点击翻译按钮获取中文摘要

One of the most difficult problems in the study of human perception involves the ability of observers to correctly interpret patterns of image shading. The light that reflects from a visible surface toward the point of observation is influenced by several factors. These include the surface shape, the pattern of illumination, and the nature of the surface material. The problem for perceptual theory is to explain how it is possible to tease apart these separate influences in order to make reliable judgments about each one. This project is designed to investigate how this is achieved by human observers. There are many potential benefits of this research. The accurate perception of shape and material properties are important for many basic human functions, such as grasping objects or judging whether visible surfaces are appropriate to afford safe locomotion. A better understanding of these processes may help to facilitate training to avoid injury due to falling. It could also lead to improved 3D visualization in medical devices, or aids for visually impaired individuals that can detect slippery spots or irregular terrain features in the path of locomotion. The first specific aim of this proposal is to measure how the perception of 3D shape is influenced by changes in surface material (e.g., glass, wax or plastic) or the pattern of illumination. This will be achieved by measuring local 3D orientation at numerous probe points on an object, from which it possible to compute a best fitting surface that is most consistent with an observer's judgments. The second specific aim is to measure how the perceptual categorization of surface materials is influenced by changes in 3D shape and the pattern of illumination. This will be achieved using a novel categorization task in which observers are shown a shaded object, and are required to adjust a set of sliders to indicate their confidence that the depicted material is one of several possible material categories. The stimuli for these experiments will be created using state-of-the-art computer graphics to produce physically accurate simulations of real materials. The results of this research will greatly improve our current understanding of the specific computational mechanisms by which patterns of image shading provide information for perceiving the 3D shapes of objects and the identification of their material properties.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.