Introduction Computer Vision Fall 2018 Columbia University - - PowerPoint PPT Presentation

▶
introduction
SMART_READER_LITE
LIVE PREVIEW

Introduction Computer Vision Fall 2018 Columbia University - - PowerPoint PPT Presentation

Nov 18, 2023 297 likes •1.17k views

Introduction Computer Vision Fall 2018 Columbia University Cameras everywhere Also scary times What is vision? What does it mean, to see? The plain man's answer (and Aristotle's, too) would be, to know what is where by looking.

slide-1
SLIDE 1

Introduction

Computer Vision Fall 2018 Columbia University

slide-2
SLIDE 2

Cameras everywhere

slide-3
SLIDE 3

Also scary times

slide-4
SLIDE 4

What is vision?

“What does it mean, to see? The plain man's answer (and Aristotle's, too) would be, to know what is where by looking.” — David Marr, 1982

1945 - 1980 (35 years old)

slide-5
SLIDE 5

Computational Photography

slide-6
SLIDE 6

Biometrics

1984

  • "the most recognized

photograph” in the history of the National Geographic magazine

  • No one knew her identity…
slide-7
SLIDE 7

Biometrics

1984 2002

slide-8
SLIDE 8

Optical Character Recognition

slide-9
SLIDE 9

Security and Tracking

“The work was painstaking and mind-numbing: One agent watched the same segment of video 400 times. The goal was to construct a timeline of images, following possible suspects as they moved along the sidewalks. It took a couple of days” Washington Post

slide-10
SLIDE 10

Health

slide-11
SLIDE 11

Gaming

slide-12
SLIDE 12

Shopping

slide-13
SLIDE 13

Special Effects

slide-14
SLIDE 14

Visual Search

slide-15
SLIDE 15

Self-driving Cars

slide-16
SLIDE 16

Space Exploration

slide-17
SLIDE 17

Augmented Reality

slide-18
SLIDE 18

Worldwide Insight

Walmart in Wichita, Kansas

slide-19
SLIDE 19

What is vision?

Slide credit: Kristen Grauman

slide-20
SLIDE 20

Image Formation

Slide credit: Steve Seitz

Object Film

slide-21
SLIDE 21

Image Formation

Add a barrier to block off most of the rays

Slide credit: Steve Seitz

Object Film Barrier

slide-22
SLIDE 22

Representing Digital Images

Slide credit: Deva Ramanan

slide-23
SLIDE 23

Representing Digital Images

Slide credit: Deva Ramanan

slide-24
SLIDE 24

Representing Color Images

R G B

Color images, RGB color space

slide-25
SLIDE 25
slide-26
SLIDE 26

Illumination

“Neither Autopilot nor the driver noticed the white side of the tractor trailer against a brightly lit sky, so the brake was not applied.” — Tesla Company Blog

Slide credit: S. Ullman

slide-27
SLIDE 27

Occlusion

René Magritte, 1957

slide-28
SLIDE 28

Class Variation

Slide credit: Antonio Torralba

slide-29
SLIDE 29

Clutter and Camouflage

slide-30
SLIDE 30

Color

slide-31
SLIDE 31

Motion

Slide credit: S. Lazebnik

slide-32
SLIDE 32

Ill-posed Problem

slide-33
SLIDE 33

Ill-posed Problem

slide-34
SLIDE 34

Ill-posed Problem

slide-35
SLIDE 35

Cambrian Explosion

Time

slide-36
SLIDE 36

Cambrian Explosion

"The Cambrian Explosion is triggered by the sudden evolution of vision,” which set off an evolutionary arms race where animals either evolved or died. — Andrew Parker

Slide credit: Fei-Fei Li

slide-37
SLIDE 37

Evolution of Biological Eye

slide-38
SLIDE 38

A quick experiment
 Animals or Not?

You will see a mask, then image, then mask. What do you see?

Slide credit: Jia Deng

slide-39
SLIDE 39
slide-40
SLIDE 40
slide-41
SLIDE 41
slide-42
SLIDE 42
slide-43
SLIDE 43
slide-44
SLIDE 44
slide-45
SLIDE 45
slide-46
SLIDE 46

Slide credit: Jia Deng

slide-47
SLIDE 47
slide-48
SLIDE 48

Thorpe, et al. Nature, 1996

150$ms$!!$

slide-49
SLIDE 49

Why not build a brain?

About 1/3rd of the brain is devoted to visual processing

slide-50
SLIDE 50

Do we have the hardware?

parallel neurons

1011 108

serial transistors

slide-51
SLIDE 51

We don’t know the software

slide-52
SLIDE 52
slide-53
SLIDE 53

Adelson Illusion

slide-54
SLIDE 54

Illusionary Motion

slide-55
SLIDE 55

Scale Ambiguity

slide-56
SLIDE 56

The Ames Room

slide-57
SLIDE 57

(Effect used in Lord of the Rings)

The Ames Room

slide-58
SLIDE 58
slide-59
SLIDE 59

Heider-Simmel Illusion

slide-60
SLIDE 60

What objects are here?

Slide credit: Rob Fergus and Antonio Torralba

slide-61
SLIDE 61

Context

Slide credit: Rob Fergus and Antonio Torralba

slide-62
SLIDE 62

Context

Slide credit: Fei-Fei Li, Rob Fergus and Antonio Torralba

slide-63
SLIDE 63

Tool 1: Physics and Geometry

slide-64
SLIDE 64

Tool 2: Data and Learning

slide-65
SLIDE 65

Two Extremes of Vision

Slide credit: Aude Oliva

slide-66
SLIDE 66

Evolution of Vision Datasets

Slide credit: Aude Oliva

Created here in 1996

slide-67
SLIDE 67

Course Information

Computer Vision Fall 2018 Columbia University

slide-68
SLIDE 68

About Me

UC Irvine

slide-69
SLIDE 69

About Me

UC Irvine MIT

slide-70
SLIDE 70

About Me

UC Irvine Google MIT

slide-71
SLIDE 71

About Me

UC Irvine Google MIT Columbia

slide-72
SLIDE 72

What about you?

  • Major?
  • Year?
  • Research area?
slide-73
SLIDE 73

Staff and Office Hours

  • Carl Vondrick


Office Hours: Monday 4:30pm to 5:30pm
 CSB 502 (temporary)

  • TAs:
  • Oscar: TBA
  • Xiaoning: Monday, 5-6pm, CS TA Room
  • Bo: Tuesday, 3-4pm, CS TA Room
  • James: TBA
  • Luc: TBA
slide-74
SLIDE 74

FAQ: Can you add me?

  • We’re at capacity: 110 people enrolled
  • 200 people on wait list
  • If you don’t plan to take class, please drop soon
slide-75
SLIDE 75

FAQ: Do I need to know C?

  • No. The problem sets will use Python.
  • Familiarity with linear algebra and calculus will be helpful

but not required.

slide-76
SLIDE 76

FAQ: How to contact you?

  • No emails — please use Piazza
  • You can send private messages on Piazza
  • Course staff goes offline 7pm to 10am and weekends
slide-77
SLIDE 77
slide-78
SLIDE 78
slide-79
SLIDE 79

Grading

  • 60% Problem Sets
  • 40% Final Project
  • No exams or quizzes
slide-80
SLIDE 80

Problem Sets

  • 5 problem sets, equally weighted
  • Turn in via CourseWorks before class starts. Submit both

PDF writeup and code online.

  • One problem set may be a week late. No other

extensions.

  • Solutions available during TA office hours.
  • Done individually, but you can have high-level discussion

in pairs. Write up assignments individually

slide-81
SLIDE 81

Final Project

  • Individually or pairs (recommended)
  • Final poster presentations: Dec 5 and Dec 10
  • 4 page report in CVPR format
  • Suggested projects and grading rubric to be announced
slide-82
SLIDE 82

Academic Honesty

  • Academic dishonesty may result in…
  • You fail course.
  • We refer your case to the Dean’s office.
slide-83
SLIDE 83

Readings (Optional)

http://szeliski.org/Book/

slide-84
SLIDE 84

New Course

  • Feedback appreciated.
  • Please let us know if something works or not!
slide-85
SLIDE 85
slide-86
SLIDE 86

Next Class: Linear Filters