Computer%20vision%20class,%20%20fast-forward

1
http//www.ai.mit.edu/courses/6.801/
2
(No Transcript)
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
Vision

• What does it mean, to see? to know what is
  where by looking.
• How to discover from images what is present in
  the world, where things are, what actions are
  taking place.

from Marr, 1982
9
Computer vision class, fast-forward
10
Images and image formation
11
Cameras, lenses, and sensors

• Pinhole cameras
• Lenses
• Projection models
• Geometric camera parameters

From Computer Vision, Forsyth and Ponce,
Prentice-Hall, 2002.
12
Radiometry
Wolfgang Lucht
http//geography.bu.edu/brdf/brdfexpl.html
13
Color
From Foundations of Vision, by Brian Wandell,
Sinauer Assoc., 1995
14
Low-level vision
15
Image filtering

• Review of linear systems, convolution
• Bandpass filter-based image representations
• Probabilistic models for images

Image
Oriented, multi-scale representation
16
Non-linear filtering, and applications
17
Models of texture
18
Shape from shading
Shape from shading, Horn and Brooks, MIT Press,
1986
19
Bayesian framework for vision
Gahan Wilsons Still Weird, Forge, 1994
20
Bayesian framework for vision
Coincidental appearance of face profile in rock?
http//www.cs.dartmouth.edu/whites/old_man.html
21
Bayesian framework for vision
Coincidental appearance of faces in rock?
http//bensguide.gpo.gov/3-5/symbols/print/mountru
shmore.html
22
Edge detection and optical flow
M. Heath, S. Sarkar, T. Sanocki, and K.W. Bowyer,
"A Robust Visual Method for Assessing the
Relative Performance of Edge-Detection
Algorithms" IEEE Transactions on Pattern Analysis
and Machine Intelligence, Vol. 19, No. 12,
December 1997, pp. 1338-1359.
23
Edge detection and optical flow
True optical flow
Images
in "Statistical Theories of the Brain", edited
by R. Rao, B. Olshausen, and M. Lewicki, MIT
Press, 2001
Estimated optical flow
24
Learning and vision
25
Statistical classifiers

• MIT Media Lab face localization results.
• Applications database search, human machine
  interaction, video conferencing.

26
Support vector machines and boosting
Large-margin classifier
www.support-vector.net/nello.html
27
Support vector machines and boosting
The kernel trick
www.support-vector.net/nello.html
28
Applications

• Computer vision for computer games

29
Computer vision applicationsas ocean-going
vessels
30
Game Decathlete
31
Optical-flow-based Decathlete figure motion
analysis
32
Decathlete 100m hurdles
33
Decathlete javelin throw
34
Decathlete javelin throw
35
video
36
Nintendo Game Boy Camera

• Several million sold (most of any digital
  camera). Imaging chip is Mitsubishi Electrics
  Artificial Retina CMOS detector.

37
video
38
The course, in broad categories

• Images and image formation
• Geometry
• Low-level vision
• High-level vision
• Implementations and applications