de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Poster

Colour and lightness of a surface seen behind a transparent filter

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons84892

D'Zmura,  M
Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons84173

Rinner,  O
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons83926

Gegenfurtner,  K
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

D'Zmura, M., Rinner, O., & Gegenfurtner, K. (1998). Colour and lightness of a surface seen behind a transparent filter. Poster presented at 21st European Conference on Visual Perception, Oxford, UK.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-E813-A
Abstract
We measured how the colour and lightness of a surface seen to lie behind a transparent filter depend on filter properties. A convergence model suggests that a filter's transformation of chromatic information from underlying surfaces is interpreted as a convergence in colour space (D'Zmura, Colantoni, Knoblauch, and Laget, 1997 Perception 26 471 - 492). Such a convergence is described by a transparency parameter alpha and by a colour that acts as the centre of convergence. We used an asymmetric matching task to test the model. In computer-graphic simulation, observers adjusted the colour of a surface seen behind a transparent colour filter in order to match the colour of a surface seen in plain view. We varied the lightness and chromatic properties of both the surface to be matched and the transparent filter. We found that the convergence model fitted the matching data nearly as well as a more general affine transformation model, even though the latter has many more parameters (twelve) than the former (four). Linear transformation, translation, and Von Kries scaling models all performed poorly. The convergence model of transparency is a general model of colour constancy. It can account for shifts in colour, such as those caused by changing the spectral properties of illumination, and can also account for shifts in contrast, like those caused by fog or by change in the spatial distribution of illumination.