Intro To COMPUTER VISION

Introduction to Computer Vision
Module introduction
Andrea Cavallaro
Todays agenda
Computer vision: what for?

The Marr model
Design of computer vision applications
Aims of the module
Objectives
Module structure
Coursework
Assessment
Computer vision
Make computers understand images and video
Who is here?
What are the objects in this room?
Is there an exit anywhere?
Does this icecream cone look good?
Computer vision
Safety
Health
Comfort
Fun
Security
Access
The Marr model of computer vision

information extracted
from the images
Understanding
symbolic data
Analysis
processsed images
Processing
input images/video
Three main steps

Processing
initial image manipulation
noise removal, image improvement, object contours enhancement
Analysis
extraction of data of non-pictorial format
number and location of objects in the scene, contours, areas, shape
Understanding
actual interpretation of the visual scene
extraction of the visual information
what are these people doing?
is this patient sick?
how many stars are in this galaxy?
Todays agenda

The Marr model
Aims of the module
Objectives
Module structure
Coursework
Assessment

Need to understand
tasks
users
context of use
Design requires the following inputs
technical
creative
human factors
user experience
Aims of this module

To provide an understanding of the most relevant
theoretical and empirical knowledge of
computer vision principles,
from its foundations to applications
To cover the fundamentals necessary to design
and develop a wide range of computer vision solutions
To study basic mathematical models used to analyze
and interpret the content in images and videos
To discuss important topics related to
real-world computer vision applications
Objectives
To understand fundamental tasks involved
in processing video data
To become familiar with the most important techniques
to perform computer-assisted video interpretation
To become familiar with basic filter design, feature
extraction, object representation, region and
object segmentation
To have an appreciation of the demands of computer
vision and challenges involved in their implementation
Book
Computer Vision:
Algorithms and Applications
Richard Szeliski
Springer
ISBN 978-1-84882-934-3 2011
http://szeliski.org/Book/
Module structure
Module structure
Lectures
Thursdays 9 11 am & 12 - 1 pm
If you miss one lecture catch up before the next one
Labs
Thursdays 1-3 pm
Exercises on computer vision tasks

Implementations can be done in any programming language
Two-hour supervision in Lab with Assistant Tutor
Labs exercises and Coursework are linked
Communication and advice

Message board is the preferred medium
Problems with coursework, labs, etc.
1. Contact the Assistant Tutor during lab hours
2. Use the Message Board
Intranet
http://qmplus.qmul.ac.uk/course/view.php?id=1914
Todays agenda

The Marr model
Aims of the module
Objectives
Module structure
Coursework
Assessment
Coursework
Implementation of computer vision tasks
The final results will be a stand alone computer programme
that can run in any PC in the School (ITL)
The design of a graphical user interface is desired,
but not mandatory
Time available
Lab time
Individual study hours
Coursework: time
Time Management
important and challenging skill that you have to acquire
in your University career
in your professional life
to be largely managed on your own

to develop an ability to schedule tasks so they can be completed
to the required standard
in the allotted time
to set aside time for coursework throughout the set period,

not just the week before submission
One of the principal reasons why some students complain about the amount
of coursework is that they are finding it hard to schedule the work properly,
leading to last minute frantic efforts to get the work in on time.
Assessment of the coursework

2 coursework submissions
first submission: code + report
second submission: code + report
2 in-lab assessments
just after the submissions
during the lab hours (in-lab assessments)
Report
Use the provided layout (template)
Coursework: software and data

Software
You can use your favourite programming language, as long as it
is supported in the ITLs computers
The functions/procedures/classes you write for the coursework
will start with the prefix ICV_
You can use freeware software, as long as the source is
acknowledged
The software shall be commented
the comments should allow an intermediate programmer to
understand each part of the code
Data set
You will be given a small database with files containing the data
to be analysed
Assessment
Final exam:
50% of the final mark
one written paper
Coursework:
50% of the final mark
individual coursework
development of computer vision tasks
What did we learn today?

The Marr model
Aims of the module
Objectives
Module structure
Coursework
Assessment
10

Intro To COMPUTER VISION

Uploaded by

Copyright:

Available Formats

Intro To COMPUTER VISION

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Intro To COMPUTER VISION

Uploaded by

Copyright:

Available Formats

Introduction to Computer Vision

Computer vision: what for?

Introduction to Computer Vision

Introduction to Computer Vision

The Marr model of computer vision

Three main steps

Introduction to Computer Vision

Computer vision: what for?

Design of computer vision applications

Design requires the following inputs

Introduction to Computer Vision

Aims of this module

Introduction to Computer Vision

Exercises on computer vision tasks

Introduction to Computer Vision

Communication and advice

Computer vision: what for?

Introduction to Computer Vision

to be largely managed on your own

to set aside time for coursework throughout the set period,

Introduction to Computer Vision

Assessment of the coursework

Coursework: software and data

Introduction to Computer Vision

50% of the final mark

one written paper

50% of the final mark

What did we learn today?

Computer vision: what for?

Introduction to Computer Vision

You might also like