September 2011
Volume 11, Issue 11
Vision Sciences Society Annual Meeting Abstract  |   September 2011
Depth Compression and Expansion in Photographs
Author Affiliations
  • Emily A. Cooper
    Helens Wills Neuroscience Institute, University of California, Berkeley
  • Elise Piazza
    Vision Science Program, University of California, Berkeley
  • Martin S. Banks
    School of Optometry, Dept. of Psychology, Helen Wills Neuroscience Institute, Vision Science Program, UCSF/UC Berkeley Joint Graduate Group in Bioengineering, University of California, Berkeley
Journal of Vision September 2011, Vol.11, 65. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Emily A. Cooper, Elise Piazza, Martin S. Banks; Depth Compression and Expansion in Photographs. Journal of Vision 2011;11(11):65.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Photographs taken with long focal length lenses appear to be compressed in depth and those taken with short focal lengths appear expanded. A common rule in photography is to use a 50-mm lens to create a natural-looking image (i.e., neither compressed nor expanded in depth). We hypothesized that this rule is a byproduct of people's viewing habits and an inability to take viewing distance into account. The retinal images generated by a photograph and the original scene are the same when the viewer is at the photograph's center of projection (CoP). The distance of the CoP is the lens focal length multiplied by the print magnification. When viewers look at a photograph from closer or farther than the CoP, the geometric information in the retinal image specifies a scene that is respectively expanded or compressed. To test our hypothesis, we conducted an experiment examining how CoP distance, print size, and image content affect preferred viewing distance. Observers adjusted their distance from photographs with different CoP distances and print sizes until the distance was “best to view from.” In agreement with studies of television viewing, preferred distance was mostly determined by print size. There was a small effect of CoP distance, due mostly to viewing portraits. Generally, observers stood too far from photographs with short CoP distances (short focal lengths) and too close to ones with long CoP distances (long focal lengths). Coupled with an inability to take distance into account when interpreting 3D contents of the photograph, this causes perceived expansion and compression with short and long focal lengths, respectively. The effects are minimized when the focal length is 50–70mm, which is consistent with common practice in photography. Our results offer guidelines for creating images – photographs, computer-generated, perspective paintings – with the desired perceived depth.

NSF Grant BCS-0617701, Department of Defense through the National Defense Science & Engineering Graduate Fellowship Program, NIH Training Grant T32 EY007043. 

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.