CSCI 363
Home | | Schedule | | Assignments | | Lecture Notes
Instructor
Constance Royden
Office: Haberlin 308
Extension: 2472
Email: croyden@holycross.edu
Office Hours: Mon & Wed 3:00 - 4:00 p.m., Tues. 1:00 - 3:00 p.m., or by appointment.
Lecture times
Mon, Wed, Fri 2:00 - 2:50 p.m.
Course description
Computational vision is an important field within computer science that
combines information from mathematics, computer science and neuroscience to
understand how a machine or biological visual system works. The study of
computer vision has been an active field and has generated many exciting
results that increase our understanding of the complex and remarkable task of
interpreting the world around us from visual information. This course will
bring together results from the field of computer vision and from the field of
biological vision and examine how different approaches can give us the ability
to recognize objects or interpret motion from a 2 dimensional image that
changes over time.
The course offers an introduction to the algorithms underlying machine and biological visual systems. It examines the processes involved in converting a 2-dimensional image to a 3-D representation of the physical world. Computational models of visual processing will be compared to physiological and psychophysical results from human and other biological visual systems. The topics covered include: edge detection, stereopsis, motion computation, shape from shading, color and object recognition.
Prerequisites
The prerequisite for this class is CSCI 132, Data Structures, and Calculus.
You should feel comfortable with basic C++ programming, and with basic calculus
concepts such as derivatives and integrals.
In addition you should be prepared to learn some new mathematical techniques. For certain concepts in computational vision, we'll need to use some mathematical methods that may be new to you, such as convolutions and fourier analysis. We will learn these methods in class.
Textbook and Readings
Vision Science. Photons to phenomenology. by Palmer.
This textbook covers a broad range of vision science from biological, psychological and
computational perspectives.
The readings in the textbook will be supplemented with readings that will be handed out. You will be expected to read the assigned material before class and be prepared to discuss the concepts from the readings. In addition, you will be asked to hand in written answers to selected questions on the readings.
Course requirements
Exams: There will be two midterm exams during the semester. They will be
on the following dates:
There will be a final exam during the regular exam period
Homework Assignments: There will be homework assignments periodically during the semester. These will involve some programming using Matlab or C++ and some paper and pencil problem solving.
Course Project: Students will work in groups to complete a course project. The project will involve researching an area of vision science and writing a paper on your findings. The research may include implementing a simple vision algorithm if you choose to do so. Projects will be presented to the class at the end of the semester.
Likely Grading (This is still being determined)
Late Policy
Assignments are due at the beginning of class on the assigned due date.
Late assignments will be marked down 10% for each day late. That is, assignments
turned in after the time they are due will be marked down 10%, assignments turned
between 24 and 48 hours after the due date will be marked down 20%, and so on. The
penalty will be determined when the assignment is physically transferred to the
instructor or signed in by any Math/CS faculty member or the department secretary.
Late work will not be accepted after the graded assignment is returned to the class.
Collaboration Policy
You are allowed to discuss strategies for solving homework problems
with other students, however any work you turn in must be your own work (i.e.
you may not simply copy another student's answers and turn them in as your own).
In addition you must clearly indicate the names of any students you work with on
each assignment.
You may consult public literature (books, articles, etc) for information, but you must cite each source of ideas you adopt.
Please familiarize yourself with the department's policy on Academic Integrity.
Home | | Schedule | | Assignments | | Lecture Notes
Constance Royden--croyden@holycross.edu
Computer Science 363--Computational Vision
Last Modified: June 26, 2016
Page Expires: August 25, 2017