Paper Summaries Any takers? Color Assignments Projects - - PDF document

1 Color

Paper Summaries

• Any takers?

Assignments

• Checkpoint 4

– Grading complete

• Checkpoint 5

– Due today

• Checkpoint 6

– Given today

• Renderman

– Due Nov 4th – Getting distributions on mycourses – Server name: see news item on mycourses

Projects

• Approx 17 projects
• Listing of projects now on Web
• Presentation schedule

– Presentations (20 min max) – Last 3 classes (week 10 + finals week) – Sign up

• Email me with 1st , 2nd , 3rd choices
• First come first served.
• Mid-quarter report due last week

Finals date

• For last day of presentations
• Friday, November 18th
• 12:30pm – 2:30pm
• 70-1445
• Times on SCHEDULE

Photography and Light

pho•tog•ra•phy, n., the process or art of producing images of objects by the action of light on a sensitized surface, e.g., a film in a camera.

Photography = writing with light

2

Computer Graphics as Virtual Photography

camera (captures light) synthetic image camera model (focuses simulated lighting)

processing

photo processing tone reproduction real scene 3D models Photography: Computer Graphics: Photographic print

Light -- What it is

• Electromagnetic radiation

power induction heating radio waves infrared ultra violet x-rays gamma rays 1016 1014 1010 108 1012 106 102 1 10-2 10-4 10-6 10-8 Wavelength (nm) 104 visible light secondary cosmic rays

Red

• range

yellow green blue violet 700 nm 650 nm 600 nm 550 nm 450 nm 400 nm

Light and Color

• “Indeed rays, properly expressed, are

not colored”

• - Sir Isaac Newton
• I.e., light rays are not colored; we

perceive them as colored!

Light - Color

• Color is the perceptual response to

light of wavelengths 400 - 700 nm hitting the retina.

• Spectral power distributions exist in

the physical world but color exists only in the eye and brain, e.g., there is no real white light!

Light – Spectral Density Functions (SDF)

• AKA spectral power distributions
• Describes the distribution of the strengths of

light at given wavelengths emitted from a source.

Light - Color

Black Body Radiators

Spectrum resulting from heating a standard

“body” to a given temperature

Plank’s formula:

) 1 ( ) , (

/ 5 1

2

− =

λ

λ λ

T c

e c T M

16 1

10 7418 . 3

−

× = c

2 2

10 4388 . 1

−

× = c

3

Light - Color

• Black Body Radiators and daylight

– Daylight from the sun & total sky (5000K - 7000K) – D65 - Average daylight (6504K) – Daylight w/occluded sun (> 7000K) – Daylight from sun alone (< 5000K)

Light -- Color

• Black Body Radiators and other light sources

Light - Color

• Not all lighting sources have smooth SDFs

Light -- Color

Light Filters

Absorbs light at given

wavelengths

Allows light at other

wavelengths through

Using filters

Actual SDF is determined

by multiply SDF of light by SDF of filter wavelength by wavelength. SDF for a filter

Light and color

• Absorption

– Material can absorb light on a wavelength by wavelength basis – Responsible for object color

Light and color

• The “color” of an object we see is a function of:

– Spectral qualities of the material being viewed:

• Absorption
• Reflection
• Diffraction
• Etc.

– Spectral qualities of the illuminating light.

4

Light and color

• Color appearance applets

http://www.cs.rit.edu/~ncs/color/a_spectr.html http://www.cs.brown.edu/exploratories/freeSoftware /repository/edu/brown/cs/exploratories/applets/spe ctrum/reflection_guide.html

Light and Color

• Color is the perceptual response to light of

wavelengths 400 - 700 nm hitting the retina.

• When rendering, spectrum must be

sampled.

• Color vision is inherently trichromatic.

Light and Color

• CIE Experiments – used X,Y,Z values to quantify

chromatic characteristics of color stimuli

Light and Color

• Color matching applet

http://www.cs.rit.edu/~ncs/color/a_game.html

• There are lots of color spaces and most of the time

we can convert between them, but not always.

Light and Color

• CIE RGB curves
• 20
• 10

10 20 30 40 3 7 5 4 4 2 5 4 5 4 7 5 5 5 2 5 5 5 5 7 5 6 6 2 5 6 5 6 7 5 7 7 2 5 7 5 Wavelength R G B

Light and Color

• CIE XYZ color matching curves

50 100 150 200 3 7 5 4 4 2 5 4 5 4 7 5 5 5 2 5 5 5 5 7 5 6 6 2 5 6 5 6 7 5 7 7 2 5 7 5 Wavelength y x z

5

Light and Color

• Chromaticity coordinates (X,Y,Z have no

perceptual correlates, although Y is luminance, and x and z provide hue information)

Z Y X X x + + = Z Y X Y y + + = Z Y X Z z + + =

1 = + + z y x

Chromaticity Coordinates

• often given in xyY
• xy give the

chromaticity

• Y gives brightness

x y

Light and Color

• RGB (or any primary set) can be

determined from XYZ

– Need chromaticies of primaries and white point.

• Primaries generally determined by device.
• RGB values are incomplete without

specification of primaries & white point.

Light and Color

• sRGB

– Standard proposed by Microsoft and HP – Based on ITU-R 709.BT – It is a lighting model for “many” CRTs

Z Y X B Z Y X G Z Y X R 0570 . 1 2040 . 0556 . 0416 . 8760 . 1 9692 . 4986 . 5374 . 1 2410 . 3 + − = + + − = − − =

Light and Color

• sRGB

http://www.cs.rit.edu/~ncs/color/a_chroma.html

Light and Color

• Other color spaces

– HSV (hue-saturation-value) – CMYK (printing) – CIELAB / CIELUV (perceptual)

• Why does CG use RGB?

– Convenience

6

Light -- Color

• Full spectral renderers are hard to find

– Expensive in time and memory – Most renderers specify color using RGB triplet (red, green, blue) – For accuracy, must convert from SDF to RGB

• Full spectral rendering going on at RIT

Munsell Color Lab (Mark Fairchild)

Light - Color

• Converting from SDF to RGB.
• 20
• 10

10 20 30 40 3 7 5 4 4 2 5 4 5 4 7 5 5 5 2 5 5 5 5 7 5 6 6 2 5 6 5 6 7 5 7 7 2 5 7 5 Wavelength R G B

Light - Color

• Converting from SDF (S) to RGB

∫

=

λ

λ λ λ d S r R ) ( ) (

∫

=

λ

λ λ λ d S g G ) ( ) (

∫

=

λ

λ λ λ d S b B ) ( ) (

Light - Color

• Converting from SDF to RGB

* * * = R = G = B Based on how “average” eye works

Light - Color

• Problems with direct conversion to RGB

– Negative values – Which RGB? (may not match RGB of monitor)

• Solution: Use XYZ space

Light - Color

• Converting SDF to XYZ

50 100 150 200 3 7 5 4 4 2 5 4 5 4 7 5 5 5 2 5 5 5 5 7 5 6 6 2 5 6 5 6 7 5 7 7 2 5 7 5 Wavelength y x z

7

Light - Color

• Converting SDF to XYZ

∫

=

λ

λ λ λ d S x X ) ( ) (

∫

=

λ

λ λ λ d S y Y ) ( ) (

∫

=

λ

λ λ λ d S z Z ) ( ) (

Light - Color

• Converting from SDF to XYZ

* * * = X = Y = Z

Light - Color

• Problems with using XYZ

– Non-intuitive – Not an abundance of XYZ renderers

• Good if you are starting with SDFs
• Good as an interchange space

Light - Color

• Converting XYZ -> RGB

– need definition of your primaries (R, G, B) in terms of XYZ coordinates

Z r Y r X r R

Z Y X

+ + = Z g Y g X g G

Z Y X

+ + = Z b Y b X b B

Z Y X

+ + =

Light - Color

• Converting XYZ -> RGB

– Construct the following matrix:

= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡

Z Z Z Y Y Y X X X

b g r b g r b g r M Light - Color

• Converting from XYZ->RGB

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ Z Y X B G R

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡

Z Z Z Y Y Y X X X

b g r b g r b g r

8

Light - Color

• Converting from RBG -> XYZ

– Invert matrix – For any color (R, G, B) we can calculate (X,Y,Z)

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ B G R Z Y X

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡

Z Z Z Y Y Y X X X

b g r b g r b g r

• 1

Light - Color

• White Point

– Chromaticity of point (1, 1, 1)

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ 1 1 1 Z Y X

⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡

Z Z Z Y Y Y X X X

b g r b g r b g r

• 1

White Point

Wikipedia

Light -- Color

• A SDF will result in a single RGB triplet.
• However, an RGB triplet can be the result of many

SDFs.

• Metamer -- Separate SDFs that produce the same

sensation of color.

• Interestingly though, reflectance and transmission

reactions are not necessarily the same, nor need the response be the same under different light sources!

Light - Color

• Example of Metamers (perceived the same)

Light - Color

• Metamers applet

http://www.cs.brown.edu/exploratories/freeSoftware /repository/edu/brown/cs/exploratories/applets/spe ctrum/metamers_guide.html

9

Light – Color Summary

• In order to produce photorealistic images, we

really need to know a lot about light, color and perception!

• Physical world -- light expressed using SDFs

– Standards based on physics – Filters

• Perceptual World - color triplets

– RGB / XYZ – Metamers

Break The 2nd half

• Color Devices
• More color perception

Display Devices

• Two Problems to be addressed by display

models

– Gamma

• Luminances from simulation are on a linear scale.

Most display devices are non linear

– Gamut

• Chromaticities calculated may not be reproducible
• n a given device due to a limited color gamut.

Display Devices

– Luminances from observer model are based on a linear scale. – Most display devices are non linear.

Display Devices

• CRTs respond non-linearly to voltage
• This non-linearity is described by gamma
• where

– Ld is the actual display luminance – Ldmax is the maximum display luminance – V is the voltage [0,1]

γ

) (

maxV

L L

d d =

10

Display Devices

• CRTs are non-linear

Sample input to monitor Graph of input Output from monitor Graph of output

Display Devices

• Gamma correction

Sample input to monitor Graph of input Gamma correction Graph of Gamma correction Output from monitor Graph of output

Display Devices

• Most displays/video cards now have gamma

control as part of their OS.

– If we can correct so that gamma is 1.0 then, getting using Ldmax from specs, the voltage V is given by

max d d

L L V =

1/γ

Display Devices

• Gamut

– Range of chromaticities reproducible by a device

Display Devices

• Different Devices have different gamuts

Display Devices

• Perceptual color spaces

– CIELAB (L*a*b*) – Distances between color values corresponds to difference in perception – Computed from X,Y,Z values and X,Y,Z of a reference white.

11

CIELAB

• L*a*b*

– L* represents luminance – a* represents position between green and red – b* represents position between green and blue

• Perceptual color space.
• Standardized by CIE

CIELAB

where Xn, Yn and Zn are the CIE XYZ tristimulus values

• f the reference white point.

for

• therwise

Display Devices

Handling out of gamut colors

http://www.cs.rit.edu/~ ncs/color/a_spaces.html

Display Devices

• Display Models must address

– Gamma / non-linearity of device – Gamut

• Usually dealt with by Color Management

Systems.

• Questions?

Viewing Environment

• Viewing environment can affect image perception
• Adaptation
• The process by which the visual mechanism adjusts to the conditions

under which the eyes are exposed to radiant energy.

– General brightness adaptation

• Adjustments in response to the overall level of stimulus exposed

– Lateral brightness adaptation

• Adjustments in response due to stimulus in adjacent areas of the retina

– Chromatic adaptation

• Adjustments in response to the average chromaticity in the stimulus.

General Brightness Adaptation

• Note also…differences in acuity

[Ferwerda 1996]

12

Lateral Brightness / Chromatic Adaptation

Chromatic Adaptation Applet Lateral Brightness (Surround Effect) Surround demos

Color and Image Appearance Models

• Color Appearance Models

– Used to predict color appearance – Accounts for changes in viewing enviornment

• Color of illuminant
• Illumination level
• Surround luminance
• Image Appearance Models

– Incorporates spatial and temporal properties of human vision. – iCAM (Munsell Color Science Lab)

Color

• Color is perceptual
• Depends upon

– Spectral Density Functions – Devices – Environment

• Questions?