Inverse Rendering Methods for Hardware-Accelerated Display of Parameterized Image Spaces

1 / 85

About This Presentation

Title:

Inverse Rendering Methods for Hardware-Accelerated Display of Parameterized Image Spaces

Description:

Inverse Rendering Methods for HardwareAccelerated Display of Parameterized Image Spaces –

Number of Views:25

Avg rating:3.0/5.0

Slides: 86

Provided by: ResearchM53

Category:

more less

Transcript and Presenter's Notes

Title: Inverse Rendering Methods for Hardware-Accelerated Display of Parameterized Image Spaces

1
Inverse Rendering Methods forHardware-Accelerated
Display of Parameterized Image Spaces
Ziyad S. Hakura
2
Real-time rendering of Parameterized Image
Spaces with Photorealistic Image Quality
Object Motion
Viewpoint Position
3

Animation time parameter
Viewpoint parameter along circle

4
(No Transcript)
5
(No Transcript)
6
Cockpit Lighting
Day light
Night sky
7
Interactive Toy Story

Limited viewpoint motion
Head motion parallax puts the user in the scene
Character parameters e.g. happiness/sadness
Rose98, Gleicher98, Popovic99

8
Parameterized Image Spaces

Space can be 1D, 2D or more
Content author specifies parameters

Light motion
Object motion
Viewpoint position
9
(No Transcript)
10
Interactive Motion
View
Light
An interactive user is free to move anywhere in
the parameter space
11
(No Transcript)
12
Ray-Tracing
Texture-Mapping Graphics Hardware
13
Ray Tracing
Display
Eye
14
Z-Buffer Graphics Hardware
15
Pixel Fill Rate vs. Time
16
Overall Model
17
Overall Model
18
Overall Model
19
Related Work

Hardware Shading Models
Diefenbach96, Walter97, Ofek98, Udeshi99,
Cabral99,
Kautz99, Heidrich99
Image-Based Rendering (IBR)
Chen93, Levoy96, Gortler96, Debevec96, Shade98,
Miller98, Debevec98, Bastos99, Heidrich99, Wood00
Animation Compression
Guenter93, Levoy95, Agrawal95, Cohen-Or99

20
Contributions

Inverse rendering method for inferring texture
maps
Hardware-accelerated decoding of compressed
parameterized image spaces
Parameterized environment maps representation
for moving away from pre-rendered samples
Hybrid rendering for refractive objects

21
Outline

Motivation
Texture Inference
Parameterized Texture Compression
Parameterized Environment Maps
Hybrid Rendering
Conclusion

22
Consider a Single Image
p2
p1
Parameterized Image Space
Single Image
How do we represent the shading on each object?
23
Texture Mapping

3D Mesh
24
Texture Inference by Inverse Rendering

3D Mesh
2D Texture
25
Linear Hardware Model
A x b
Unknown Texture Pixels
Ray-Traced Image
HW Filter Coefficients
26
Texture Inference
x texture values
b ray-traced image
Ax b r(x)2
27
Forward Mapping Method
28
Ray-Traced
Inverse Fitted PSNR41.8dB
Forward Mapped PSNR35.4dB
29
Outline

Motivation
Texture Inference
Parameterized Texture Compression
Parameterized Environment Maps
Hybrid Rendering
Conclusion

30
Parameterized Texture Compression
Light
View
31
Why compress textures instead of images?

Textures better capture coherence
Independent of where in image object appears
Object silhouettes correctly rendered from
geometry
Viewpoint can move away from original samples
No geometric disocclusions

32
Laplacian Pyramid
33
Adaptive Pyramid
34
MPEG-Image, 3551 PSNR36.8dB
Laplacian-Texture, 3791 PSNR38.7dB
35
(No Transcript)
36
Runtime System

Decompresses texture images
Caches uncompressed textures in memory
Textures in top of pyramid likely to be re-used
Generates rendering calls to graphics system

37
(No Transcript)
38
(No Transcript)
39
Outline

Motivation
Texture Inference
Parameterized Texture Compression
Parameterized Environment Maps
Hybrid Rendering
Conclusion

40
How to handle reflective objects?
Problem Movement away from pre-rendered views
gives a pasted-on look Solution Parameterized
Environment Maps
41
Static Environment Maps (EMs)

Generated using standard techniques
Photograph a physical sphere in an environment
Render six faces of a cube from object center

42
Problem with Static EM
Ray-Traced
Static EM
Self-reflections are missing
43
ParameterizedEnvironment Maps (PEM)
44
(No Transcript)
45
Environment Map Geometry
46
Why Parameterize Environment Maps?

Captures view-dependent shading in environment
Accounts for geometric error due to approximation
of environment with simple geometry

47
Surface Light Fields Miller98,Wood00
Surface Light Field
PEM
Dense sampling over surface points
of low-resolution lumispheres
Sparse sampling over viewpoints of
high-resolution EMs
48
Layering of Paramaterized Environment Maps

Segment environment into local and distant maps
Allows different EM geometries in each layer
Supports parallax between layers

49
Segmented, Ray-Traced Images
Fresnel
EMs are inferred for each layer separately
50
Inferred EMs per Viewpoint
Distant
Local Color
Local Alpha
51
Experimental Setup

1D view space
1 separation between views
100 sampled viewpoints

52
Ray-Traced vs. PEM
Closely match local reflections like
self-reflections
53
(No Transcript)
54
Movement Away from Viewpoint Samples
Ray-Traced
PEM
55
(No Transcript)
56
Layered PEM vs. Infinite Sphere PEM
Layered PEM
Infinite Sphere PEM
57
(No Transcript)
58
(No Transcript)
59
Outline

Motivation
Texture Inference
Parameterized Texture Compression
Parameterized Environment Maps
Hybrid Rendering
Conclusion

60
How to handle refractive objects?
Problem Outgoing ray direction hard to predict
from first surface intersection
N
Refractive Path
Eye
Outgoing ray
Solution Hybrid Rendering
61
Hybrid Rendering
62
Hybrid Rendering

Greedy Ray Path Shading Model
Adaptive Tessellation
Layered, Parameterized Environment Maps

63
Greedy Ray Path Shading Model
Reflective Path
Refractive Object
N
Refractive Path
Eye
Trace two ray paths until rays exit refractive
object
64
Comparison of Shading Models
Two-term greedy ray path
Full ray tree
65
Adaptive Tessellation

Two criteria
Ray path topology
Outgoing ray distance
Consider both terms of shading model

66
Layered EMs
67
Inferred EMs
L1
L2
L3
Reflection Term
Refraction Term
Inferred Environment Maps
68
Ray-Traced vs. Hybrid
69
(No Transcript)
70
Benefit of Hybrid Renderingover Ray-Tracing