August 2014
Volume 14, Issue 10
Vision Sciences Society Annual Meeting Abstract  |   August 2014
Specularity and shape from line drawings
Author Affiliations
  • Flip Phillips
    Psychology & Neuroscience, Skidmore College
  • Julia Mazzarella
    Psychology & Neuroscience, Skidmore College
  • Pete Docter
    Pixar Animation Studios
Journal of Vision August 2014, Vol.14, 729. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Flip Phillips, Julia Mazzarella, Pete Docter; Specularity and shape from line drawings. Journal of Vision 2014;14(10):729. doi:

      Download citation file:

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

  • Supplements

Line drawings afford observers with a surprising amount of information about shape. For example, Picasso's Fragment de corps de femme (See provides an breathtakingly legible depiction of the human form that is rich with shape information from a mere four lines. Our previous work (Phillips, Casella & Gaudino 2005) as well as the work of others (Cole, Sanik & DeCarlo 2009) has investigated the nature of the underlying 3D geometric information used by artists when executing a line drawing of an object. Along with primal information like the object's boundary contour, illustrators tend to represent internal self-occlusion contours and other features largely determined by the intrinsic geometric properties of the object. It is well known that material properties that modulate the reflectance function of the object also contribute to the perception of shape. In fact, it is possible to identify objects made of glossy or shiny materials using only their specularity. This study extends our previous work on illustrators' depiction of intrinsic geometry to depiction of material properties, specifically shininess. A professional artist/illustrator depicted multiple versions of a sphere using a variety of indications of shininess. Subjects provided gauge figure adjustments for the various illustration conditions, allowing us to reconstruct the perceived shape of the depicted spheres. For all conditions, subjects underestimated the 3D curvature of the sphere. The best-fitting estimations came from stimuli with a simple highlight depiction that was roughly consistent with the curvature of the theoretical sphere. Other techniques yielded perceptual depth and curvature whose accuracy depended on the properties of the artistic indication of gloss. Geometrically consistent or plausible highlights provided more depth than those that were not.

Meeting abstract presented at VSS 2014


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.