Judgments of Size and Distance in Photographs

Classical theories of space perception posit continuous distortions of subjective space. These stand in contrast to the quantitatively and qualitatively different distortions experienced in space that is represented pictorially. Wechallenge several aspects of these theories. Comparing real-world objects with depictions of the same objects, we investigated to what extent distortions are introduced by the photographic medium. Comers of irregularly shaped buildings had to be judged in terms of the vertical dihedral angles subtended by two adjacent walls. Across all conditions, a robust effect of viewing distance was found: Buildingcomers appear to flatten out with distance. Moreover,depictions of comers produce remarkably similar results and should not receive a different theoretical treatment than do real-world scenes. The flattening of vertical angles cannot be explained by a linear distortion of the entire visual space. We suggest that, for natural scenes, compression of space is local and dependent on contextual information. Visually perceived depth in spatial arrangements has been studied within the context of two rather distinct frameworks. First, many theories of subjective space are based on the notion that perceived space is distorted in a uniform fashion and that a specific transformation can be found that describes the mapping relations between the Euclidean space of the world and our subjective space. Second, theories of spatial representation are concerned with distortions that arise in the process of representing space-mainly, the effects caused by pictorial rendition, such as projective distortions or the apparent flattening of the scene. In this study, we attempt to establish a relationship between the two frameworks by comparing judgments made in real scenes with judgments based on photographic renditions ofthe same scenes. Surprisingly, we found distortions to be very similar. We had observers judge the dihedral angles of building comers from a va-H

Section: Pictorial Effects On Perceived Visual Spacementioning

confidence: 99%

Compression of visual space in natural scenes and in their photographic counterparts

Hecht

Doorn

Koenderink

1999

“…Similarly, varying the observer's viewing distance to a picture changes egocentric and exocentric depth estimates but not estimates of object width and size (Bengston, Stergios, Ward, & Jester, 1980;O. W Smith, 1958aO. W Smith, , 1958bO.…”

mentioning

confidence: 99%

Weakening the robustness of perspective: Evidence for a modified theory of compensation in picture perception

Yang

Kubovy

1999

Viewed from the center of projection, a perspective picture presents the pictorial depth information of a scene. Knowing the center of projection, one can reconstruct the depicted scene. Assuming another viewpoint is the center of projection will cause one to reconstruct a transformed scene. Despite these transformations, we appreciate pictures from other viewpoints. The compensation hypothesis states that the visible picture surface allows observers to compensate for transformations by locating the center of projection and experiencing pictorial space from there. Weshow that observers neither completely compensate for nor experience transformations of space as geometry would predict. We propose a modified compensation hypothesis according to which different degrees of visibility of the picture surface invoke different degrees of compensation.Pictures are flat surfaces that show scenes in depth. For the visual system, pictures present the problem ofintegrating conflicting flatness and depth information. Practically, understanding how picture perception resembles or differs from perceiving real space will allow us to create and use spatial displays more effectively. In this paper, we address two problems: (1) how we perceive pictures from different viewpoints even though pictures are geometrically correct for only one viewpoint and (2) how depth information and flatness information interact in perceiving depth in pictures.A picture mimics the light from a scene to one viewpoint, called the center ofprojection. To all other viewpoints, the picture presents a geometrically transformed pictorial space. However, experience suggests that we can appreciate pictures from many viewpoints. Kubovy (1986) has called this phenomenon the robustness ofperspective. Some researchers (e.g., Goldstein, 1987;Kubovy, 1986;Pirenne, 1970; have proposed that robustness results because seeing the picture surface allows observers to compensate for these transformations ofpictorial space. That is, observers perceive the layout of pictorial space as if they were viewing the This research was supported by Grant MH 47317 to M.K., principal investigator. The authors thank Jake Mazulewicz, Anu Kesari, and Albert Downs for their able help in conducting the experiment, and Ron Simmons for building the apparatus. The authors also thank Marco Bertamini for discussions on the nature of explanations and descriptions, Hal Sedgwick for his thoughts on cross-talk, the visual world, and the visual field, Denny Proffitt for his thoughts on "compellingness" in pictures, movies at SIGGRAPH, and registered variables, and Marco Bertamini and Mukul Bhalla for their suggestions, which greatly improved this paper. Correspondence should be addressed to T. Yang or M. Kubovy, Department of Psychology, Gilmer Hall, University ofVirginia, Charlottesville, VA 22903-2477 (e-mail: yang@virginia.edu or kubovy@virginia.edu).picture from the center of projection. We call this explanation the compensation theory ofperspective robustness.In this paper, we present two versio...

“…In some cases, authors find that perception is essentially compatible with projective geometry and that the distortions predicted by the transformed virtual space are indeed perceived (Bengston, Stergios, Ward, & Jester, 1980;Busey, Brady, & Cutting, 1990;McGreevy & Ellis, 1986;Rosinski, Mulholland, Degelman, & Farber, 1980;Smith & Gruber, 1958). However, other evidence suggests that perception does not always conform to the geometry of polar projection (Buseyet al, 1990;Farber & Rosinski, 1978;Goldstein, 1987;Smith, 1958;Wallach & Marshall, 1986). Since the discovery and documentation of the geometric rules for producing polar projection drawings in the Renaissance, many have noticed that for some polar projection drawings, departures from the correct vantage point do not have serious consequences for the drawings' visual impression (Edgerton, 1975;Kubovy, 1986;Pirenne, 1970;Veltman, 1987;White, 1967).…”

mentioning

confidence: 99%

Angular subtense effects on perception of polar and parallel projections of cubes

Nicholls

Kennedy

1993

Many authors contend that the perception of 2-D drawings of a 3-D object is governed by polar projective geometry. A problem for this position is that observers accept parallel projections, which are not produced with polar projective geometry, as accurate representations of 3-D objects. In Experiments 1 and 2, we used two different standards of comparison to study the perceptions of three line drawings of cubes-correct polar projections of cubes with subtenses of 15°and 35 0, and a parallel projection-at five different angular subtenses. In Experiment 1, 14 observers judged each drawing when it subtended about 35 0, 15°, 5°, 4°, and 2°in width. Subjects used an 8-point rating scale to compare each drawing with a correct polar projection of a cube subtending 35 0, viewed with the drawing subtending 150. As predicted, both polar projections had their highest ratings at their correct vantage points. Ratings for the parallel projection were highest at small angular subtenses and decreased when it subtended 35 0. These findings were supported by a second experiment in which the 15°polar projection was set at a 5°viewing angle as a standard. In Experiment 3, 15 observers compared the three drawings, viewed at a second set of angular subtenses (30°,35°,40°,45°, and 50°), with a standard, the 35°polar set at 45°. Ratings fell with increases in viewing angle, and the parallel projection was rated lowest. The results indicate that parallel projections are assessed as polar projections that are correct for objects at a small angular subtense. Furthermore, projections at a small angular subtense are robust; that is, they are acceptable over a wide range of angular subtenses. We suggest that robustness can be explained by the modest variability in the proportions of pictures of cubes subtending small angles.An important problem for the study of the perception of pictures lies in explaining how accurate impressions of three-dimensional (3-D) scenes can occur through the viewing of 2-D pictures. Here, we will examine two projection systems that have been used extensively to create 2-D pictures-polar projection and parallel projection. In polar projection drawings, lines representing edges that recede from the viewer converge on the picture surface and, if extended, would meet at vanishing points. In parallel projection drawings, lines representing parallel, receding edges are drawn as parallel on the page.Objects project light to a vantage point, creating an optic array at the point, in what is called natural perspective.