|
General Information
-
Time and Place: MCCB 1108, Mon, 7th period (1:55am-2:45pm),
Wed, 7th and 8th periods (1:55pm-2:45pm and 3:00pm-3:50pm)
-
Instructor: Prof. Jeffrey Ho (jho[AT]cise[.]ufl.edu)
Office Hours: CSE 360, Mon 8th and Wed 9th periods respectively or by appointment (to be setup via email).
-
TA: Mohsen Ali (moali[AT]cise[.]ufl.edu)
Office Hours: CSE 445, Monday (12:00pm-1:00pm)and Thursday (12:50pm-1:50pm).
-
Prerequisites: Linear Algebra, Multivariate Calculus, proficiency in MATLAB or C/C++ and familiarity with Image I/O.
Texts and Other Material
-
Recommended Text: Computer Vision - A modern approach, by D. Forsyth and J. Ponce, Prentice Hall.
The following link has more information on the textbook:
Forsyth's page on the book .
- Other References:
1. Computer Vision: Algorithms and Applications by R. Szeliski. This book (draft) is available online.
2. Introductory Techniques for 3D Computer Vision, by E. Trucco and A. Verri, Publisher: Prentice Hall.
3. Fourier Transforms and its Applications, Bracewell, McGraw Hill Co.
4. Other Material: Papers from the following IEEE TPAMI, CVIU (Academic Press, formerly: CVGIP), the AI Journal, and IJCV.
Syllabus
-
Image Formation: Monocular imaging system; Orthographic and
Perspective Projections; Camera model and Camera calibration; Binocular
imaging systems; 3D image sensing (range sensing with laser ranging,
grid coding etc.)
-
Low-level Vision: Basic image processing (continuous and
discrete images), Edges and edge finding, stereo vision,
Regularization, Shape from X, Optic flow and it.s computation, Motion
analysis (computation of motion parameters and structure).
-
Shape Segmentation and Representation: Simple segmentation
techniques in 2D and 3D: Deformable curves and surfaces a.k.a. .snakes.
and associated numerical methods. Snakes for tracking and Kalman
snakes. Normalized Cuts and Graph Cuts. 2D (implicit and explicit
functions, boundaries: Fourier/Wavelet descriptors; regions: Texture
description using co-occurrence matrices, Medial axis, quadtrees etc.)
and 3D shape representation (surface based:implicit and explicit
functions, Gauss map and its differential; and volume based: Octrees,
deformable solids etc.) techniques, Multi-resolution representations
(Laplacian pyramid and wavelet basis).
-
High-level Vision: Simple object recognition methods in 2D and
3D: Various criteria for imageimage or shape-shape matching. Principal
Component Analysis (PCA) and building priors for recognition.
Grading
- Homeworks and Programs: 30%. NO late assignments.
Homeworks/programs will be assigned biweekly. All programs must
be written in MATLAB. Your program output will invariably involve
processed image display and MATLAB provides a convenient platform for doing image processing. It is available on
all departmental workstations/PCs, and you are expected to learn how to use MATLAB on your own through various online sources.
- Two Midterms: 25% each; There is NO makeup policy unless there are medical reasons and
the student will be required to show a letter from an MD as proof.
First Midterm will be given approximately at the middle of the semester and the second will
be in the last week of classes.
- Programming Project: 20%
Projects may be done individually or in teams of at most two. The project will involve
two aspects namely, (a) a two page initial description, which will not be graded but must
be submitted within the first 10 weeks of class for approval; a list of projects, depending
on enrollment, may be handed out in class but considerable freedom will be given to the
students in choosing their own projects, and (b) a final report (no more than 5 pages of
text not including the images and references) that will be due on the last class day. Project
demonstrations may be arranged at a mutually convenient time on an individual/team basis.
Other Policies:
-
Academic Honesty: See
http://www.dso.ufl.edu/judicial/honestybrochure.htm for Academic
Honesty Guidelines. All academic dishonesty cases will be handled
through the University of Florida Honor Court procedures as documented
by the office of Student Services, P202 Peabody Hall. You may contact
them at 392-1261 for a "Student Judicial Process: Guide for Students"
pamphlet.
- You are allowed to discuss hwk and programing problems with
your colleagues in class but keep in mind that you MUST hand in your
own solutions which must be distinct from those you consulted with.
- Cell Phones:
Your cell phones MUST be switched off before you come to the class.
- Students with Disabilities: Students requesting
classroom accommodation must first register with the Dean of Students
Office. The Dean of Students Office will provide documentation to the
student who must then provide this documentation to the Instructor when
requesting accommodation.
|