CS781 - Colour for Computer Graphics - Winter 2009

Lecture 10

Visual Response to Light

Colour matching functions depends on something done by humans,

• namely, colour matching
• How do they do it?

1. The Eye

2. The Retina

3. Neural Processing

4. Opponent colours theory

Ganglion Cell & Cortical Physiology

Ganglion cells we understand pretty well

1. Centre-surround organization
   Pretty well every combination
   • red-green/red+green X centre/surround X inhibitory centre
   • blue centre/red+green surround
2. Only colour opponency, no spatial opponency
   • blue/red+green

Cortical cells

Hubel & Weisel (Nobel prize for this), "There is a feedforward network: location -> edges -> motion -> etc."

• The Livingstone reference gives their story.

More recently,

1. Charlie Michael, "Every cell ever predicted by a psychophysicist has been found in the cortex, and then some."
2. Peter Lennie, "Thirteen different channels of processing handle colour in one way or another."
3. Samir Zeki, "I have found cells sensitive to individual wavelengths."

Psychophysical phenomena

How to do experimentation.

1. Reddish-green and yellowish-blue don't exist
   • reddish-yellow, reddish blue, greenish-yellow, greenish-blue do exist
   • relationship to hue circle
2. Unique hues
   • yellow, red, blue: simple
   • green: bimodal
3. Flicker photometry
   • luminance channel
     • highest frequency
     • linear
   • red-green channel
   • yellow-blue channel
4. Minimally distinct border
5. Colour naming
   • Berlin & Kay
6. Unusual properties of yellow-blue axis

Neural Theories

Other interesting observations

1. Basic colours
   • extension of colour naming
   • Boynton's experiment
2. Cardinal colours
   • based on adaptation
3. Simultaneous contrast

Return to:

• Bill Cowan's Winter 2009 cs781 lecture note page
• Bill Cowan's Winter 2009 cs781 page
• Bill Cowan's cs781 page
• Bill Cowan's teaching page
• Bill Cowan's home page