The Perception and Content of Cast Shadows: An Interdisciplinary Review

Abstract

Recently, psychologists have turned their attention to the study of cast shadows and demonstrated that the human perceptual system values information from shadows very highly in the perception of spatial qualities, sometimes to the detriment of other cues. However with some notable and recent exceptions, computer vision systems treat cast shadows not as signal but as noise. This paper provides a concise yet comprehensive review of the literature on cast shadow perception from across the cognitive sciences, including the theoretical information available, the perception of shadows in human and machine vision, and the ways in which shadows can be used.

Keywords:

• shadows
• perception
• spatial reasoning
• spatial perception

Notes

¹There is a large sub-field of vision research that deals with shape-from-shading whereby an object's self-shadowing is used to determine its shape, such as Kriegman and Belhumeur (1998). This line of research, however, does not take into account cast shadows. For more detail on shape from shading, see the recent review paper by Durou, Falcone, and Sagona (2008).

²Although a very worthwhile mention here has to go to Chinese shadow puppetry. While this art, strictly speaking, is concerned with using and not depicting shadows, it has been around for millennia.

³That is, the physics necessary to cast a copycat shadow is richer than standard physics, and so cannot be a simplified physics as Cavanagh suggests.

⁴ http://gizmodo.com/5582218/what-directx-11-is-and-what-it-means-to-you

⁵Optical contact is the place where an object is connected to the background in a 2D projection of a 3D scene.

⁶It is worth mentioning that even if shadows are discounted, there is no evidence that this discounting may affect functions other than object identification. The information in shadows could (at least in principle) still be used for depth estimation.

⁷In video, the reference pixel is at the same spatial location but from the background model, in a still image, the reference pixel is a neighbor.

⁸This rule is not true in all cases; we must assume that the authors were not considering objects with holes

⁹That is, in the first few hundred milliseconds of processing that does not involve stimulus-specific knowledge.