2/20/24
Computer Vision
Course
School of Informations and
Communication Technology
1
2/20/24
Computer Vision
Chapter 1: Introduction
Computer Vision Group
School of Information and Communications Technology
Contents
• What is computer vision?
• Concepts and definitions
• Level of vision (low, middle and
high)
• Related fields
• Application areas
CV Group – School of Information and Communication and Technology
4
4
2
2/20/24
What is computer vision?
Human vision
Image or video Sensing (device) Interpreting
Interpretations
Computer vision
From CS131 course “computer vision”,
Prof. Fei-Fei Li, Stanford 'Vision'Lab
The goal of computer vision
• To bridge the gap between pixels and “meaning”
Source: S. Narasimhan
What we see What a computer sees
26-Sep-17 6
6
3
2/20/24
What kind of information can we extract from an image?
• Metric 3D information
• Semantic information
Vision as measurement device
Structure from motion Reconstruction from
Internet photo collections
NASA Mars Rover
Pollefeys et al. Goesele et al.
4
2/20/24
Vision as a source of semantic information
sky amusement park Objects
Activities
Scenes
Locations
The Wicked
Cedar Point Text / writing
Twister
Faces
Ferris Gestures
wheel Motions
ride ride Emotions…
12 E
Slide credit: Kristen Grauman
Lake Erie water
tree ride
tree people waiting in line
people sitting on ride
carousel
tree umbrellas
deck maxair
bench tree pedestrians
What is computer vision?
• Computer vision
• Is an interdisciplinary scientific field that deals with how computers
can be made to gain high-level understanding from digital images or
videos.
• From the perspective of engineering, it seeks to automate tasks that
the human visual system can do.
• Computer vision tasks include
• methods for acquiring, processing, analyzing and understanding digital
images,
• and extraction of high-dimensional data from the real world in order
to produce numerical or symbolic information, e.g., in the forms of
decisions. (Wikipedia).
10
10
5
2/20/24
What is computer vision?
• The two definitions of CV can be defined as a scientific field
that extracts information out of digital images.
• Another way to define CV is through its applications.
• Computer vision is building algorithms that can understand the
content of images and use it for other application
What kind of scene?
Where are the cars?
How far is the building?
…
11
11
Levels of vision
• Low-level Vision: Image Formation, Acquisition,
Image Processing
• Image formation studies the forward process of
producing images and videos
• Image acquisition:
• A digital image is produced by several image sensors.
• Depending on the type of sensor, the resulting image
data is an ordinary 2D image, a 3D volume, or an
image sequence.
• Image processing focuses on 2D image data
processing using point operators such as contrast
enhancement, filtering (local operations), noise
reduction, image transforms. Image processing is
considered as pre-processing that is usually
necessary to process the image data for CV
applications.
• Work with image as a matrix
• Input: image è output: image
12
12
6
2/20/24
Levels of vision
• Middle-level Vision: Feature, Image matching
• Feature extraction: Image features at various levels of
complexity are extracted from the image data. Examples
of such features: Edges, ridges, lines, texture, shape …
• Image matching
• Image segmentation
13
13
3D urban modeling
Bing maps, Google Streetview
Source: S. Seitz
14
14
7
2/20/24
Levels of vision
• High-level Vision: High-level vision is to infer the semantics,
for example, object recognition and scene understanding.
• Several application topics:
• Object recognition (classification), Identification
• Detection
• Motion analysis
• Scene reconstruction; 3D reconstruction
• Image-based rendering ….
15
15
Related fields
Computer vision at the intersection of multiple scientific fields
16
16
8
2/20/24
Related fields
Source: From EECS 432-Advanced Computer Vision, Northwestern University
17
17
Related fields
• Machine Learning: “The field of study that gives computers the
ability to learn without being explicitly programmed.” – Arthur
Samuel
• Artificial intelligence and computer vision share other topics such
as pattern recognition and learning techniques.
• Computer vision - Deep learning: Artificial Neural Networks with
many layers (CNN: Convolutional Neural Network )
18
18
9
2/20/24
Applications areas
Robotics Application
• Localization-determine robot location automatically
• Navigation
• Obstacles avoidance
• Assembly peg − in − hole, welding, painting
• Manipulation e. g. PUMA robot manipulator
• Human Robot Interaction HRI: Intelligent robotics to
interact with and serve people
19
19
Applications areas
Security Application
• Biometrics iris, fingerprint, face recognition
• Surveillance-detecting certain suspicious activities or
behaviors
• …
Face recognition systems
Fingerprint scanners
Source: from S. Seitz
20
20
10
2/20/24
Examples - Face Detection
Source: from S. Seitz
21
21
Examples of Computer Vision
Slide from Vicente Ordonez
22
22
11
2/20/24
Applications areas
Medicine Application
• Classification and detection e. g.
• 2D/3D segmentation
• 3D human organ reconstruction MRI or ultrasound
• Vision-guided robotics surgery
• …
Slide from Jason Lawrence
23
23
Applications areas
Industrial Automation Application
• Industrial inspection defect detection
• Barcode and package label reading
• Object sorting
• Document understanding e. g. OCR
• …
Transportation Application
• Autonomous vehicle
• Safety, e.g., driver vigilance monitoring
• …
24
24
12
2/20/24
Facebook's suggestion
• Understand the image
25
25
Examples- Toys and Robots
26
26
13
2/20/24
Examples - Optical character recognition (OCR)
License plate readers
Digit recognition,(AT&T labs) http://en.wikipedia.org/wiki/Automatic_number_pl
ate_recognition
Source: from S. Seitz
27
27
THANK YOU !
CV Group – School of Information and Communication and Technology 28
28
14
