Steve Seitz - PowerPoint PPT Presentation

1 / 36
About This Presentation
Title:

Steve Seitz

Description:

Steve Seitz – PowerPoint PPT presentation

Number of Views:30
Avg rating:3.0/5.0
Slides: 37
Provided by: grailCsWa
Category:
Tags: idem | seitz | steve

less

Transcript and Presenter's Notes

Title: Steve Seitz


1
3D PhotographyBeyond Perspective
  • Steve Seitz
  • Dept. Computer Science Eng.
  • University of Washington

2
Perspective Projection
  • Humans evolved with perspective eyes
  • Capture light along rays that converge at a
    single point

3
Cameras also evolved with perspective optics
  • Optimized for humans, not computers!

4
A non-perspective image
5
(No Transcript)
6
The Blue Marble, NASA satellite image composite
7
The Blue Marble, NASA satellite image composite
8
The Blue Marble, NASA satellite image composite
9
Cyberware Scanner
10
Print Gallery, by M.C. Escher, 1956
11
Panorama from Disneys 1940 film Pinocchio (from
Wood et al., SIGGRAPH 1997)
12
(No Transcript)
13
Whats an Image?
  • An image is any 2D subset of rays in space
  • Actually, the light energy flowing along these
    rays
  • Any 2D slice of the plenoptic function

perspective image
14
Non-perspective Imaging
  • Issues
  • What other types of images are possible?
  • Which images are useful?
  • How can we capture these images?

15
Path Images
16
Path Images
17
Path Images
y
x
t
18
Path Images
y
x
t
19
Path Images
  • Pushbroom images
  • satellite
  • Bolles et al. 87
  • EPI
  • Tsuji et al. 92
  • omni-directional image
  • Peleg et al. 97
  • manifold mosaic
  • Radamacher Bishop 99
  • MCOP
  • . . .

y
x
t
20
Linear Path
  • Video Cube Demo
  • application by Michael Cohen et al., Microsoft

input
21
Circular Path
22
Circular Path
input
cyclographs
video cube
23
  • What are these images good for?
  • Applications to computer graphics, computer
    vision?

24
Circular Stereo
  • Ishiguro, Yamamoto, Tsuji, 92
  • Peleg and Ben-Ezra, 99
  • Shum, Kalai, Seitz, 99
  • Nayar and Karmarkar, 00

I
25
Stereo Panorama
Disparity map result
26
Stereo Panorama
Dark--close, light--far
27
Circular Stereo
  • Advantages
  • 360 degree scene reconstruction
  • Uniform accuracy, optimal

28
Pushbroom Stereo
29
Stereo Cyclographs
30
(No Transcript)
31
(No Transcript)
32
Stereo Cyclograph Reconstruction
  • Computed from two cyclograph images
  • Using unmodified stereo matcher Zitnick Kanade

33
Stereo Path Images
Yes
  • Do these all produce stereo pairs?
  • two images with horizontal parallax

34
Stereo Path Images
No
  • How about this path?
  • Yes if the camera path is a conic
  • line, circle, ellipse, parabola, hyperbola
  • Must capture rays lying on doubly-ruled quadrics
  • Padja 2001, Seitz 2001

35
Stereo Parabolic Panoramas
36
Beyond Perspective
  • Cameras for humans, not machines!
  • Need to rethink cameras
  • Image should suit the task
  • Future cameras will evolve like CPUs
  • First task-specific cameras
  • Then programmable cameras
  • FPGA ? programmable camera arrays
  • Thanks to Jiwon Kim, Michael Cohen
Write a Comment
User Comments (0)
About PowerShow.com