CS 436 / CS 5310 :: Computer Vision Spring 2015
Instructor
Office
Office Extension
Email
Office Hours
Zehra Shah
9-119, SSE
8196 call Zulfiqar Naseem Malik (8313) if not contactable
[email protected]
TBA
Course Information
Credit Hours
Course Type
Class Meeting Times
Venue
Website
CS 436 / CS 5310 :: Computer Vision
3 credit hours
Elective for CS graduate students
TBA
TBA
TBA
Teaching Assistant
Office
Phone
Email
Office Hours
TBA
TBA
Course Description
This course gives a broad overview of the field of computer vision, laying the foundations for advanced
graduate level classes and research work in vision. This course will be conducted with an application
perspective. Therefore students will be expected to implement several techniques learnt in the lectures. A
good calculus, linear algebra and programming background is expected for this class. Knowledge of
probability and random variables is also needed to understand the ideas presented in some modules.
The nature of the field of Computer Vision is such that it combines and integrates ideas from several
different areas, including statistics, pattern recognition, machine intelligence, decision theory and image
processing. Therefore, in an introductory class, it is not possible to cover each of these aspects in depth.
Instead, the focus of this course is on breadth, presenting several different techniques and systems in
moderate detail, so as to familiarize the student with the Computer Vision area in general, and to present
some specific examples of Computer Vision systems.
Textbooks
There is no required text. The following reference texts are useful.
Computer Vision: Algorithms and Applications: This is the draft of a textbook recently written
by Richard Szeliski, one of the leading researchers in the area, available in PDF form at
http://szeliski.org/Book/
Introductory Techniques for 3D Computer Vision: This text, by Emanuel Trucco and
Alessandro Verri, is very useful, especially for topics related to geometry.
CS 436 / CS 5310 :: Computer Vision Spring 2015
Pre-requisites
Linear Algebra, CS 200: Introduction to Programming
Basic knowledge of programming, linear algebra, probability and calculus is required for this course.
Learning Outcomes
The broad goals of this course are:
1. Acquire a broad overview of the theory and applications of Computer Vision.
2. Understand and appreciate the difficulty of developing computer vision software and its reasons.
3. Acquire an understanding of basic computer vision techniques and mathematical models in the
areas of the geometry of image formation, transformations, feature detection, computation of 3D
structure, motion estimation, registration and recognition.
4. Understand the mathematical basis of computer vision techniques, including a few robust
estimation techniques which are commonly used in computer vision research.
5. Develop skills to write (and debug) programs to solve computer vision problems, through solving
several programming assignments.
6. Learn about some current hot problems in computer vision research and the approaches that are
being pursued.
By the end of the semester, a student should have acquired the following skills:
1.
2.
3.
4.
Explain some successful applications of computer vision algorithms and how they work.
Recognize the difficulty in solving the general image understanding problem.
Write programs to solve basic computer vision problems using MATLAB.
Understand the mathematical basis of several computer vision techniques in the areas of image
formation, transformations, feature detection, motion estimation, stereo, structure from motion
and others.
5. Read and understand a research paper of moderate difficulty in the area of computer vision.
6. Formulate an approach to a computer vision problem, implement it, debug it, and then suggest
improvements.
7. Be poised to undertake further graduate study and research in this area.
Grading Policies
The distribution of grades of each instrument is given in the table. The
following specific policies also apply:
Instrument
Assignments
Quizzes
Midterm
Final
Weight
30%
10%
25%
35%
One in every five quizzes (rounded off to nearest integer) will be
dropped.
Late submissions of assignments will be penalized at 5n % per 24
hrs past the submission deadline, where n is the number of the
assignment that the student has submitted late. If a portion of the
assignment is late, the student is encouraged to submit the completed part on time for full credit,
and the remaining part late, on which the grade penalty will be imposed.
CS 436 / CS 5310 :: Computer Vision Spring 2015
Policies for co-located course CS436 :: Computer Vision Fundamentals
A version of this course is also being offered to undergraduate students as CS436. The course outline
remains the same and the lectures will take place jointly. However, the undergraduate students will not be
tested on certain topics in the exams and their homework deliverables will be less.
Honor Code
We expect all LUMS students to act honestly at all times. Cheating and plagiarism will not be tolerated.
Resorting to unfair means may result in an F in the course and expulsion from LUMS, even for first-time
offenders.
Some examples of what constitutes cheating and plagiarism are listed below, though this is not an
exhaustive list:
Copying code from someone elses screen or notes
Providing your own written code to another student
Letting another student type solutions for you on the machine you are working on, for assignment
submission
Looking at someone elses answer sheet during a quiz or an exam
Downloading code from the internet or any other source and submitting portions of it as your own.
Writing code, whether on a computer or a sheet of paper, for someone elses assignment and providing
it to him or her.
Emailing your own code to someone else.
Debugging someone elses code.
Some examples of what does not constitute cheating and plagiarism
Answering questions on the web forum
Discussing project ideas with another student
Helping other students understand concepts
If in doubt, discuss with your instructor.
CS 436 / CS 5310 :: Computer Vision Spring 2015
Tentative Course Plan
Topic
Introduction
Course Introduction, policies, who should take this course
Overview of Computer Vision
Why are computer vision problems hard?
Examples of successful computer vision applications
Overview of course topics
Geometric Transformations and Camera Models
2D transformations
3D transformations
3D 2D transformations
Camera Models
Camera Calibration
Feature Detection
Edge Detection
Corner Detection
Fitting Lines to Data Points
Dealing with noisy data and multiple lines
Midterm Exam
Feature Based Alignment
2D feature based alignment
3D alignment
Pose Estimation
Dense Motion Estimation and Image Stitching
Optical Flow
Pyramids
Parametric Methods for Image Alignment
Structure from Motion
Rigid SFM (Factorization Method)
Non Rigid SFM
Bundle Adjustment
Stereo
Basic Formulation
Epipolar Constraint
Estimation of Fundamental Matrix
Tracking (tentative)
Lectures
1-2
Reading
Szeliski Ch 1
3-10
Szeliski Ch 2
11-14
Szeliski Ch 4
Trucco Ch 4-5
15
16-17
Szeliski Ch 6
18-20
Szeliski Ch 8-9
21-24
Szeliski Ch 7
25-26
Trucco Ch 7-8
27-28
Szeliski Ch 14
