Fundamentals of Computer

Vision
Assist. Prof. Özge ÖZTİMUR KARADAĞ
Department of Computer Engineering
ALKÜ
12 Feb Introduction to Computer Vision
19 Feb Learning Based Computer Vision - I
26 Feb Learning Based Computer Vision - II
4 March Image Features - I
11 March Image Features - II
18 March Object Recognition
25 March Classification – Neural Networks
30 March – 7 April Midterm
8 April Deep Learning
15 April Deep Learning – Architectures
22 April Deep Learning – Transfer Learning
29 April Semantic Segmentation
6 May Action Recognition
13 May Project Presentations – I
20 May Project Presentations – II
27 May Review – Application Examples
Reference Books
• Available online for free
• Computer Vision: Algorithms and Applications, Richard Szeliski
http://szeliski.org/Book/
• Computer Vision: A Modern Approach, David Forsyth
http://luthuli.cs.uiuc.edu/~daf/CV2E-site/cv2eindex.html
Evaluation
• Midterm %30
• Project : %30
• Final %40
What is Computer Vision?
• What does it deal with?
• What is its relation with image processing?
In Image Processing
• We dealt with image’s
• formation,
• representation,
• enhancement,
• segmentation

To remind you shortly:

Image Formation
Representation of the Image
• Digital image:
• a 2d f function.
• Consists of pixels.

Figür M.J.Black
Filtering
Filtering
Edge Detection
Image Segmentation

Figures from: A. Erdem

Goal in Computer Vision
• Extract ‘meaning’ from image pixels.
• Human are good at it!
• But actually, our visual system can easily deceive us .
Human vision is superbly efficient

Potter, Biederman, etc. 1970s

Checker Shadow İllusion
Visual Illusions
• Ponzo Illusion
The relation of computer vision
with other fields
 Biology
Psychology
Neuroscience
Physics optics Cognitive
sciences

Image Algorithms,
processing
Computer theory,… Computer
Science
Vision Systems,
Speech, NLP architecture, …

Robotics Information
retrieval
Engineering Machine
learning

Mathematics

Juan Carlos Niebles ve Ranjay Krishna sunumundan alınmıştır.

What is (computer) vision?

Image (or video) Sensing device Interpreting device Interpretations

garden, spring,
bridge, water,
trees, flower,
green, etc.

Juan Carlos Niebles ve Ranjay Krishna sunumundan alınmıştır.

The goal of computer vision
• To bridge the gap between pixels and “meaning”

Source: S. Narasimhan
What we see What a computer sees
Juan Carlos Niebles ve Ranjay Krishna sunumundan alınmıştır.
Change Blindness

Rensink, O’regan, Simon, etc.

Juan Carlos Niebles ve Ranjay Krishna sunumundan alınmıştır.

Change Blindness

Rensink, O’regan, Simon, etc.

Juan Carlos Niebles ve Ranjay Krishna sunumundan alınmıştır.

Sample applications of computer
vision
Neural Style Transfer
[Gatys et al. 2015]
CycleGAN [Zhu et al. 2017]
 amusement park
sky Objects
Activities
Scenes Vision as a
The Wicked Cedar Point Locations
Twister Text / writing source of
Faces
Gestures semantic
ride Ferris wheel Motions
Emotions… information
ride
12 E
Lake Erie water
ride
tree

tree

Slide credit: Kristen Grauman

people waiting in line

people sitting on ride

umbrellas
tree maxair
carousel
deck
bench tree pedestrians
3D urban modeling

Bing maps, Google Streetview

Source: S. Seitz
Face detection

• Many digital cameras now detect faces

• Canon, Sony, Fuji, …

Source: S. Seitz
Face recognition: Apple iPhoto software

http://www.apple.com/ilife/iphoto/
Biometrics

How the Afghan Girl was Identified by Her Iris Patterns

Source: S. Seitz
Biometrics

Face recognition systems now beginning

Fingerprint scanners on
to appear more widely
many new laptops, iphone X introduced face recognition
other devices
Optical character recognition (OCR)
Technology to convert scanned docs to text
• If you have a scanner, it probably came with OCR software

Digit recognition, AT&T labs License plate readers

http://en.wikipedia.org/wiki/Automatic_number_plate_recognition

Source: S. Seitz
Google maps: Annotate all houses
and streets

Goodfellow et al. 2014

Toys and Robots
Vision in supermarkets

LaneHawk by EvolutionRobotics
“A smart camera is flush-mounted in the checkout lane, continuously watching for items.
When an item is detected and recognized, the cashier verifies the quantity of items that
were found under the basket, and continues to close the transaction. The item can remain
under the basket, and with LaneHawk, you are assured to get paid for it… “
Source: S. Seitz
Vision-based interaction (and games)

Sony EyeToy

Microsoft’s Kinect

Assistive technologies
Source: S. Seitz
Augmented Reality
And more…