![Page 1: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/1.jpg)
CSCI 5561: Computer Vision, Prof. Paul Schrater, Spring 2005
Shading, Shadows, Lights, and shapefrom shading
![Page 2: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/2.jpg)
n̂θ
δA
δI
αzf
Surface Radiance and Image Irradiance
α
22 )cos/(cos
)cos/(cos
ααδ
αθδ
fI
zA
=2
coscos
=
fz
IA
θα
δδ
Same solid angle
PinholeCameraModel
![Page 3: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/3.jpg)
n̂δA
δI
αzf
d
Surface Radiance and Image Irradiance
α
απ
ααπ 3
2
2
2
cos4)cos/(
cos4
==Ωzd
zd
Solid angle subtended by the lens, as seen by the patch δA
θαπ
δθδδ coscos4
cos 32
=Ω=zdALALP
Power from patch δA through the lens
Ω
απ
θαπ
δδ
δδ 4
23
2
cos4
coscos4
=
==fdL
zd
IAL
IPE
Thus, we conclude
![Page 4: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/4.jpg)
n̂δA
δI
αzf
d
Surface Radiance and Image Irradiance
αΩ
απ
θαπ
δδ
δδ 4
23
2
cos4
coscos4
=
==fdL
zd
IAL
IPE
€
E ∝Le Image intensity isproportional to ExidentRadiance
![Page 5: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/5.jpg)
Radiance emitted by point sources• small, distant sphere
radius ε and uniformradiance E, which is faraway and subtends asolid angle of about
€
Ω = πεd
2
€
Le = ρd x( ) Li x,ω( )cosθidωΩ
∫= ρd x( ) Lidω cosθi
Ω
∫
≈ ρd x( ) εr(x)
2
E cosθi
= kρd x( )cosθi
r x( )2
![Page 6: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/6.jpg)
Standard nearby point source model• N is the surface normal
• rho is diffuse albedo
• S is source vector - a vectorfrom x to the source, whoselength is the intensity term– works because a dot-product
is basically a cosine
€
ρd x( ) N x( ) •S x( )r x( )2
![Page 7: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/7.jpg)
Standard distant point source model• Issue: nearby point source
gets bigger if one gets closer– the sun doesn’t for any
reasonable binding of closer
• Assume that all points in themodel are close to eachother with respect to thedistance to the source.Then the source vectordoesn’t vary much, and thedistance doesn’t vary mucheither, and we can roll theconstants together to get:
€
ρd x( ) N x( )• Sd x( )( )
![Page 8: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/8.jpg)
Line sources
radiosity due to line source varies with inverse distance, if the source is long enough.
For a finite line source centeredon the plane, there is spatialvariation across the surface
![Page 9: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/9.jpg)
Area sources• Examples: diffuser boxes,
white walls.
• The radiosity at a point dueto an area source isobtained by adding up thecontribution over the sectionof view hemispheresubtended by the source– change variables and add up
over the source
![Page 10: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/10.jpg)
Area Source Shadows
![Page 11: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/11.jpg)
Shading models• Local shading model
– Surface has radiosity dueonly to sources visible ateach point
– Advantages:• often easy to manipulate,
expressions easy
• supports quite simple theoriesof how shape information canbe extracted from shading
• Global shading model– surface radiosity is due to
radiance reflected from othersurfaces as well as fromsurfaces
– Advantages:• usually very accurate
– Disadvantage:• extremely difficult to infer
anything from shading values
![Page 12: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/12.jpg)
Photometric stereo
• Assume:– a local shading model
– a set of point sources that are infinitely distant
– a set of pictures of an object, obtained in exactly the samecamera/object configuration but using different sources
– A Lambertian object (or the specular component has beenidentified and removed)
![Page 13: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/13.jpg)
Projection model for surface recovery - usually calleda Monge patch
€
z = f (x, y)
r n =∂z /∂x∂z /∂y
1
ˆ n = r n / r n €
ˆ n
![Page 14: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/14.jpg)
Image model• For each point source, we
know the source vector (byassumption). We assumewe know the scalingconstant of the linearcamera. Fold the normaland the reflectance into onevector g, and the scalingconstant and source vectorinto another Vj
• Out of shadow:
• In shadow:
€
I j(x,y) = 0
€
I j (x, y) = kB(x, y)
= kρ(x, y) N(x, y) •S j( )= g(x, y)•Vj
![Page 15: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/15.jpg)
![Page 16: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/16.jpg)
CSCI 5561: Computer Vision, Prof. Paul Schrater, Spring 2005
Surface normals
![Page 17: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/17.jpg)
CSCI 5561: Computer Vision, Prof. Paul Schrater, Spring 2005
€
l1 =
− 200
€
l2 =
200
Surface NormalLight directionsDefined relative to local surface patch
![Page 18: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/18.jpg)
Basic Equation
ValuesOf images atPixel (x,y)
Matrix stackOf light sourcesFor each of the nimages
Unknown
Problem- Nonlinear due to shadowing
n x 3n x 1
€
I1(x,y)I2(x,y)
L
In (x,y)
= max
V1T
V1T
L
VnT
g(x,y),r 0
. .
etc
€
I1(x,y)
€
I2(x,y)
€
In (x,y)
.
![Page 19: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/19.jpg)
Dealing with shadows
€
I12(x, y)I22(x, y)..
In2(x, y)
=
I1(x, y) 0 .. 00 I2(x, y) .. .... .. .. 00 .. 0 In(x, y)
V1T
V2T
..Vn
T
g(x, y)
Known Known Known Unknown
€
r b = Ar x General form:
n x 3n x 1
For each x,y point
![Page 20: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/20.jpg)
Recovering normal and reflectance• Given sufficient sources, we can solve the previous
equation (most likely need a least squares solution) for g(x, y) = ρ(x,y) N(x, y)• Recall that N(x, y) is the unit normal• This means that ρ(x,y) is the magnitude of g(x, y)• This yields a check
– If the magnitude of g(x, y) is greater than 1, there’s aproblem
• And N(x, y) = g(x, y) / ρ(x,y)
![Page 21: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/21.jpg)
Example figures
![Page 22: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/22.jpg)
Recovered reflectance
![Page 23: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/23.jpg)
Recovered normal field
![Page 24: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/24.jpg)
Recovering a surface from normals - 1• Recall the surface is written
as
• This means the normal hasthe form:
• If we write the known vectorg as
• Then we obtain values forthe partial derivatives of thesurface:
€
(x, y, f (x, y))
€
g(x, y) =
g1(x, y)g2 (x, y)g3(x, y)
€
fx (x, y) = g1(x, y) g3(x, y)( )
€
fy(x, y) = g2(x, y) g3(x, y)( )
€
N(x, y) =1
fx2 + fy
2 +1
− fx− fy1
![Page 25: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/25.jpg)
Recovering a surface from normals - 2• Recall that mixed second partials
are equal --- this gives us acheck. We must have:
(or they should be similar, at least)
• We can now recover thesurface height at any pointby integration along somepath, e.g.
€
∂ g1(x, y) g3(x, y)( )∂y
=
∂ g2(x, y) g3(x, y)( )∂x
€
f (x, y) = fx (s, y)ds0
x
∫ +
fy (x, t)dt0
y
∫ + c
![Page 26: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/26.jpg)
Surface recovered by integration
![Page 27: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/27.jpg)
Shape from Shading with ambiguity
What to do if light source unknown?
Only one image?
Ans: Use prior knowledge in the form of constraints(e.g. regularization methods, Bayesian methods), takeadvantage of the habits (regularities) of images.
![Page 28: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/28.jpg)
Horn’s approach
• Want to estimate scene parameters (surface slopes fx(x,y) and fy(x,y) atevery image position, (x,y).
• Have a rendering function that takes you from some given set of sceneparameters to observation data (e.g. r(x,y) n(x,y) gives image intensity for any(x,y)).
• Could try to find the parameters fx(x,y) & fy(x,y) that minimize thedifference from the observations I(x,y).
• But the problem is “ill-posed”, or underspecified from that constraint alone.So add-in additional requirements that the scene parameters must satisfy (thesurface slopes fx(x,y) & fy(x,y) must be smooth at every point).
![Page 29: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/29.jpg)
RegularizationFor each normal, compute the distance fromthe normal to its neighbors:
€
Prior/regularizer
s(i, j) =r n i + k, j + l( ) −
r n (i, j)( )2
k={−1,1}∑l={−1,1}∑
Intensity Error
r(i, j) = I i, j( ) − Ipred (i, j)( )2
Err = r(i, j) + λs(i, j)i, j∑
![Page 30: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/30.jpg)
CSCI 5561: Computer Vision, Prof. Paul Schrater, Spring 2005
Shape from Shading Ambiguity
![Page 31: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/31.jpg)
CSCI 5561: Computer Vision, Prof. Paul Schrater, Spring 2005
![Page 32: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/32.jpg)
Curious Experimental Fact
• Prepare two rooms, one with white walls and whiteobjects, one with black walls and black objects
• Illuminate the black room with bright light, the whiteroom with dim light
• People can tell which is which (due to Gilchrist)
• Why? (a local shading model predicts they can’t).
![Page 33: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/33.jpg)
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
A view of awhite room,under dim light.Below, we see across-section ofthe imageintensitycorresponding tothe line drawnon the image.
![Page 34: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/34.jpg)
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
A view of a blackroom, underbright light.Below, we see across-section ofthe imageintensitycorresponding tothe line drawnon the image.
![Page 35: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/35.jpg)
What’s going on here?
• local shading model is a poor description of physicalprocesses that give rise to images– because surfaces reflect light onto one another
• This is a major nuisance; the distribution of light (inprinciple) depends on the configuration of everyradiator; big distant ones are as important as smallnearby ones (solid angle)
• The effects are easy to model
• It appears to be hard to extract information from thesemodels
![Page 36: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/36.jpg)
Interreflections - a global shading model• Other surfaces are now area sources - this yields:
• Vis(x, u) is 1 if they can see each other, 0 if they can’t
€
Radiosity at surface = Exitance + Radiosity due to other surfaces
B x( ) = E x( ) + ρd x( ) B u( )cosθi cosθsπr(x,u)2 Vis x,u( )dAu
all othersurfaces
∫
![Page 37: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/37.jpg)
What do we do about this?
• Attempt to build approximations– Ambient illumination
• Study qualitative effects– reflexes
– decreased dynamic range
– smoothing
• Try to use other information to control errors
![Page 38: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/38.jpg)
Shadows cast by a point source• A point that can’t see the
source is in shadow
• For point sources, thegeometry is simple
![Page 39: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/39.jpg)
Ambient Illumination• Two forms
– Add a constant to the radiosity at every point in the scene toaccount for brighter shadows than predicted by point sourcemodel
• Advantages: simple, easily managed (e.g. how would you changephotometric stereo?)
• Disadvantages: poor approximation (compare black and whiterooms
– Add a term at each point that depends on the size of theclear viewing hemisphere at each point (see next slide)
• Advantages: appears to be quite a good approximation, but jury isout
• Disadvantages: difficult to work with
![Page 40: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/40.jpg)
At a point inside a cube or room, the surface sees light in alldirections, so add a large term. At a point on the base of a groove,the surface sees relatively little light, so add a smaller term.
![Page 41: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/41.jpg)
Reflexes
• A characteristic feature of interreflections is little brightpatches in concave regions– Examples in following slides
– Perhaps one should detect and reason about reflexes?
– Known that artists reproduce reflexes, but often too big andin the wrong place
![Page 42: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/42.jpg)
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
At the top, geometry of asemi-circular bump on aplane; below, predictedradiosity solutions, scaledto lie on top of each other,for different albedos ofthe geometry. Whenalbedo is close to zero,shading follows a localmodel; when it is close toone, there are substantialreflexes.
![Page 43: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/43.jpg)
Radiosity observed in animage of this geometry;note the reflexes, whichare circled.
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
![Page 44: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/44.jpg)
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
At the top, geometry of agutter with triangularcross-section; below,predicted radiositysolutions, scaled to lie ontop of each other, fordifferent albedos of thegeometry. When albedois close to zero, shadingfollows a local model;when it is close to one,there are substantialreflexes.
![Page 45: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/45.jpg)
Radiosity observed in animage of this geometry;above, for a black gutterand below for a white one
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
![Page 46: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/46.jpg)
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
At the top, geometry of agutter with triangularcross-section; below,predicted radiositysolutions, scaled to lie ontop of each other, fordifferent albedos of thegeometry. When albedois close to zero, shadingfollows a local model;when it is close to one,there are substantialreflexes.
![Page 47: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/47.jpg)
Radiosity observed in animage of this geometryfor a white gutter.
Figure from “Mutual Illumination,” by D.A. Forsyth and A.P. Zisserman, Proc. CVPR, 1989, copyright 1989 IEEE
![Page 48: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/48.jpg)
Smoothing
• Interreflections smooth detail– E.g. you can’t see the pattern of a stained glass window by
looking at the floor at the base of the window; at best, you’llsee coloured blobs.
– This is because, as I move from point to point on a surface,the pattern that I see in my incoming hemisphere doesn’tchange all that much
– Implies that fast changes in the radiosity are localphenomena.
![Page 49: Shading, Shadows, Lights, and shape from shadingvision.psych.umn.edu/users/schrater/schrater_lab/courses/CompVis0… · Shading, Shadows, Lights, and shape from shading. n ˆ θ δA](https://reader030.vdocuments.net/reader030/viewer/2022040407/5eaaab9c9c5bfa135b191507/html5/thumbnails/49.jpg)
Fix a small patch near alarge radiator carrying aperiodic radiosity signal;the radiosity on the surfaceis periodic, and its amplitudefalls very fast with thefrequency of the signal. Thegeometry is illustratedabove. Below, we show agraph of amplitude as afunction of spatialfrequency, for differentinclinations of the smallpatch. This means that ifyou observe a highfrequency signal, it didn’tcome from a distant source.