Graphics and Archaeology: Interpreting the Past

Snowbird, Utah
20-23 May 2000

Panel:

Light, Colour, and Perception

Presenters:

• Kevin Cain
• Kevin Mitchell
• Michael Ashley Lopez
• Duncan Brown

Summary:

Kevin Cain showed three CG images and discussed the extremes taken to get the data correct (but not the lighting). The impacts of lighting on perception of the presented information could be significant.

Paul Debevec noted, however, that the colors in our reproduction media are way off all the tim. It was our brains that factored in the variances and mapped the resulting information to a constant interpretation. We accept things and make the necessary adjustments.

Kevin Mitchell presented a short statement on light, colour, and perception. He showed an image in greyscale versus infrared and noted that we recognize that the objects (something like a Landrover vehicle) were the same thing even though we cannot see in infrared directly. He went on further to preview some of his presentation the next day, especially to show us examples of how we'd lose information because of the color gamut limitations of our current monitor technology.

Duncan pondered how can we tell that we have a good representation? There is perception and science. One usually has to go with personal perception.

Conclusions / Issues:

• When imaging, we should help the user interpret light effectively, e.g., in medical imaging, it is common practice to provide a sighting ball in volume rendering to how show the type, orientation, and context of lighting.
• It is easy to produce images that look good - it is difficult to produce images that are correct across the correct spectrum of light.
• What is the importance of colour? What is real?
  • Michelangelo's work - what is real? When he first applied the paint? A week later? Years later? It all changed over time, so when is a piece significant?