BBM416 -- Fundamentals of Computer Vision Spring 2013

“ But one must learn to read, just as one must learn to see and learn to live. ” - Vincent Van Gogh, from Letters to Theo

course description

Computer vision aims to develop methods that enable machines to gain the ability to analyze and understand the visual inputs. This undergraduate level course is intended for introducing the fundamental topics of computer vision. In this context, we will first start from the low-level image perception aspects, such as image formation, cameras, color and continue with mid-level vision topics, such as interest point detection and local feature extraction. The course will also introduce the fundamentals of high-level vision tasks, such as face detection/recognition, object recognition and human motion analysis.

Prerequisites: linear algebra, basic knowledge of probability and statistics, good programming skills.

course logistics

Instructor: Nazli Ikizler Cinbis , nazli -at- cs.hacettepe.edu.tr

Office: 112, Office Hours: Thursdays 10:00-12:00

Lecture Hours: Fridays 09:00-11:45

Lecture Room: D8

textbooks

Not required, but strongly suggested

• Computer Vision: Algorithms and Applications, by Richard Szeliski, Springer, 2010. (Available free online)
• Computer Vision: A Modern Approach (2nd edition), by D.A. Forsyth and J. Ponce, Prentice Hall, 2011.
• Learning OpenCV, by Gary Bradski & Adrian Kaehler, O'Reilly Media, 2008.
• Pattern Classification (2nd Edition), by R.O. Duda, P.E. Hart, and D.G. Stork, Wiley-Interscience, 2000.

grading policy

• Midterm exam %30
• Project %25
• Final exam %45

Project: The project will involve designing and implementing a computer vision system towards solving a fundamental vision problem. The project can be carried out individually or in groups of two. In the context of BBM418 Computer Vision Laboratory, the students are required to complete four programming assignments.

schedule (tentative)

Date	Topic	Slides	Readings	Assignments
March 8th	Introduction to computer vision	.ppt	Szeliski 1
March 15th	Image formation, camera and color	.ppt	Szeliski 2, FP 1.1,3.1,3.3	Assignment 1 out
March 22nd	Filters, Templates and Image Pyramids	.ppt	Szeliski 3.2, FP 4
March 29th	Interest points and image features	.ppt	Szeliski 4.1, FP 5	Assignment 1 due, Assignment 2 out
April 5th	Edge Detection, line fitting	.ppt	Szeliski 4.1, FP 8
April 12th	Machine learning: clustering and classification overview	.ppt	Szeliski 5	Assignment 2 due
April 19th	Midterm Exam
April 26th	Instance recognition, bag-of-words models	.ppt	Szeliski 14.2, Szeliski 14.3, FP 16	Assignment 3 out
May 3rd	Object detection with sliding windows	.ppt	Szeliski 14.1, FP 17
May 10th	Motion analysis and Tracking	.ppt	FP 11	Assignment 3 due
May 17th	Human Action Recognition	.ppt		Assignment 4 out
May 24th	Image Retrieval	.ppt	FP 21
May 31th	Advanced Applications	.ppt		Final Project due
June 6th	Project presentations			Assignment 4 due

useful links

• Matlab Tutorial
• Linear Algebra Review
• Random Variables Review
• CVOnline
• OpenCV(open source computer vision library)

Academic Integrity: The students are expected to strictly adhere to the academic integrity policy of the Department, act honestly and respect the rights of the others in carrying out all academic assignments. Academic dishonesty, including cheating, fabrication and plagiarism will not be tolerated. Unless stated otherwise, all the work on the assignments must be carried out individually. While the discussions over the general concepts about the course are allowed, discussions related to a particular solution to a specific problem, (actual or pseudo) code sharing are strictly forbidden. Using assistance from the internet without providing proper citation is also considered as violation of the academic integrity. Note that, all the aforementioned violations to academic integrity policy are subject to disciplinary action.