Title: Color2Gray: Salience-Preserving Color Removal
1Color2Gray Salience-Preserving Color Removal
- Amy Gooch
- Sven Olsen
- Jack Tumblin
- Bruce Gooch
2(No Transcript)
3Problem
Volume of displayable CIE Lab Colors
4Isoluminant Colors
Color
Grayscale
5Converting to Grayscale
- In Color Space
- Linear
- Nonlinear
- In Image Space
- Pixels (RGB)
- Using colors in the image
- Different gray for different color
- Relative difference
- Using colors in the image and their position in
image space - Colors can map to same gray..
6Traditional Methods Luminance Channels
Problem can not be solved by simply switching
to a different space
CIE CAM 97
Photoshop LAB
CIE XYZ
YCrCb
7Traditional Methods
- Contrast enhancement Gamma Correction
- Doesnt help with isoluminant values
PSGray Auto Contrast
New Algorithm
8Simple Linear Mapping
Luminance Axis
9Principal Component Analysis (PCA)
10Problem with PCA
Worst case Isoluminant Colorwheel
11Non-linear mapping
12Contemporaneous Research
- Rasche et al. 2005, IEEE CGA and EG
Color Image
Luminance Only
Rasche et al.'s Method
13Goals
- Dimensionality Reduction
- From tristimulus values to single channel
- Loss of information
- Maintain salient features in color image
- Human perception
14Relative differences
Color Illusion by Lotto and Purves
http//www.lottolab.org
15Challenge 1Influence of neighboring pixels
16Challenge 2Dimension and Size Reduction
120, 120
100
0
-120, -120
17Challenge 3Many Color2Gray Solutions
Original
. . .
18Algorithm Intuition
Look at DC
. . .
i 1, j 2
luminance
Look at DC
19Algorithm Overview
- Convert to Perceptually Uniform Space
- CIE Lab
- Initialize image, g, with L channel
- For every pixel
- Compute Luminance distance
- Compute Chrominance distance
dij
- Adjust g to incorporate both luminance and
chrominance differences
20Color2Grey Algorithm
- Optimization
- min S S ( (gi - gj) - di,j )2
im
i
ji-m
21Parameters
m
Radius of neighboring pixels
a
q
- Map chromatic difference to increases or
decreases in luminance values
22m Neighborhood Size
m 2
m 16
23m Neighborhood Size
m 16
24a Chromatic variation maps to luminance
variation
a
-a
crunch(x) a tanh(x/a)
a 5
a 10
a 25
25Perceptual Distance
DLij Li - Lj
- Chrominance Distance DCij
Problem DCij is unsigned
26Map chromatic difference to increases or
decreases in luminance values
b
C2
a
-a
Color Space
C1
27Db
Color Difference Space
DC1,2
vq (cos q, sin q)
-
vq
q
Da
-Da
-
sign(DCi,j) sign(DCi,j . vq )
-Db
28q 45
q 225
Photoshop Grayscale
29q 45
q 135
q 0
Grayscale
30How to CombineChrominance and Luminance
dij
(Luminance)
DLij
31How to CombineChrominance and Luminance
(Luminance)
if DLij gt DCij
DLij
dij
DCij
(Chrominance)
32How to CombineChrominance and Luminance
DLij
if DLij gt crunch(DCij)
d (a) ij
crunch(DCij)
33How to CombineChrominance and Luminance
DLij
if DLij gt crunch(DCij)
d(a,q)ij
if DCij . nq 0
crunch(DCij)
otherwise
crunch(-DCij)
34Color2Grey Algorithm
- Optimization
- min S S ( (gi - gj) - di,j )2
im
ji-m
i
If dij DL then ideal image is g Otherwise,
selectively modulated by DCij
35Results
Photoshop Grey
Original
Color2Grey
36(No Transcript)
37Original
PhotoshopGrey
Color2Grey
38Original
PhotoshopGrey
Color2Grey
39Implementation Performance
- Image of size S x S
- O(m2 S2) or O(S4) for full neighborhood case
- 12.7s 100x100 image
- 65.6s 150x150 image
- 204.0s 200x200 image
- GPU implementation
- O(S2) ideal, really O(S3)
- 2.8s 100x100
- 9.7s 150x150
- 25.7s 200x200
Athlon 64 3200 CPU
NVIDIA GeForce GT6800
40Future Work
- Smarter
- Remove need to specify q
- New optimization function designed to match both
signed and unsigned difference terms - Image complexity measures
41Validate "Salience Preserving"
Original
PhotoshopGrey
Color2Grey
Apply Contrast Attention model by Ma and Zhang
2003
42Validate "Salience Preserving"
Original
PhotoshopGrey
Color2Grey
43Thank you
www.color2gray.info
- SIGGRAPH Reviewers
- NSF
- Helen and Robert J. Piros Fellowship
- Northwestern Graphics Group
- MidGraph2004 Participants
- especially Feng Liu
- (sorry I spelled your name wrong in the
acknowledgements)
44(No Transcript)
45Original
Color2Grey
Color2GreyColor
46Original
Color2Grey
Color2GreyColor
47Color2Grey
Original
Color2GreyColor
48Original
PhotoshopGrey
Color2Grey
49Original
Color2Grey
Color2GreyColor
50Original
PhotoshopGrey
Color2Grey
51Original
Color2Grey
Color2GreyColor
52Photoshop Grayscale
53Photoshop Grayscale
54Rasche et al.
Photoshop Grayscale
55Rasche et al.
Photoshop Grayscale
56Rasche et al.
Photoshop Grayscale
57Parameter a
a 5
a 15
a 25
a 35
a 45
a 55
a 65
a 75
a 85
a 95