CSCI3330: Fundamentals of Applied Computer Vision (Spring 2024)
Course Description
Computer vision is the enterprise of building machines that can sense and extract information
from 3D scenes. It is one of the fastest growing and exciting disciplines in today’s academia and
industry, with a wide variety of applications in robotics, video surveillance, navigation, consumer electronics,
human-computer interaction, medical imaging, remote sensing, space exploration, and so on.
This course is designed to open the doors for students who are interested in learning about the
fundamental principles and important applications of computer vision. The following topics will be covered:
cameras and image formation, image filtering, edge and corner detection, image features and matching,
image alignment, geometric camera models, binocular stereo, optical flow, radiometry, photometric stereo,
structured light, and other applications in machine vision.
Course Information
Lectures:
Tutorials:
Instructor: Prof. Jinwei Gu
TA(s): TBA
Course notes and assignments are available at Blackboard http://blackboard.cuhk.edu.hk
Reference Books
(SZ) Computer Vision: Algorithms and Applications, by Richard Szeliski. Free online
(FP) Computer Vision: A Modern Approach, by David Forsyth and Jean Ponce. Free online
(BH) Robot Vision, by Berthold K.P. Horn.
First Principles of Computer Vision, by Shree Nayar. Free online
Tentative Schedule
| Week |
Date |
Lecture |
Reading |
HW & Exam |
| 1 |
Jan 8 |
Course Introduction and Overview |
SZ Ch.1, Ch.2.3;
FP Ch.1.1; BH Ch.1, Ch.2 |
|
| Jan 9 |
Camera Model, Image Formation |
|
| 2 |
Jan 15 |
Image Sensing |
SZ Ch.3.1, Ch.3.2;
FP Ch.4.1, Ch.4.2, Ch.22 |
|
| Jan 16 |
Image Filtering (1) |
|
| 3 |
Jan 22 |
Image Filtering (2) |
SZ Ch.3.3, Ch.3.4;
FP Ch.4.3, Ch.4.4;
BH Ch.6.7 |
HW1 out |
| Jan 23 |
Camera ISP Pipeline |
|
| 4 |
Jan 29 |
Noise and HDR Imaging |
SZ Ch.7.1; FP Ch.5 |
|
| Jan 30 |
Edge and Corner Detection |
|
| 5 |
Feb 5 |
Line and Boundary Detection |
SZ Ch.7.1; FP Ch.5 |
|
| Feb 6 |
Image Features and Matching |
|
| Spring Festival Holidays |
| 6 |
Feb 19 |
Image Stitching |
SZ Ch.7.4, Ch.7.5;
FP Ch.9.1, Ch.9.3,
Ch.10.1, Ch.10.2 |
HW2 out |
| Feb 20 |
Image Segmentation |
|
| 7 |
Feb 26 |
Face Detection and Recognition |
SZ Ch.2.1, Ch.11.1;
FP Ch.1.1, Ch.1.2, Ch.1.3 |
|
| Feb 27 |
Review Session |
|
| Reading Week |
| 8 |
Mar 11 |
Midterm |
SZ Ch.8.1, Ch.8.2; FP Ch.7.1 |
Midterm |
| Mar 12 |
Geometric Camera Model |
|
| 9 |
Mar 18 |
Epipolar Geometry |
SZ Ch.12.1, Ch.12.3, Ch.12.5;
FP Ch.7.2, Ch.7.4, Ch.7.5;
BH Ch.13 |
|
| Mar 19 |
Binocular Stereo |
HW3 Out |
| 10 |
Mar 25 |
Motion and Optical Flow |
SZ Ch.9.1, Ch.9.3; BH Ch.12 |
|
| Mar 26 |
Structure from Motion |
|
| 11 |
Apr 1 |
No class (Holiday) |
SZ Ch.2.2; FP Ch.2.1, Ch.2.2;
BH Ch.9 |
|
| Apr 2 |
Radiometry |
|
| 12 |
Apr 8 |
Photometric Stereo |
SZ Ch.13.1; FP Ch.2.4;
BH Ch.10, Ch.11 |
HW4 out |
| Apr 9 |
Structured Light for 3D Capture |
|
| 13 |
Apr 15 |
Neural Networks for Computer Vision |
|
|
| Apr 16 |
Review Session |
|
Tutorial Sessions
In addition to regular lectures, this class will also have weekly
tutorial sessions. The tutorial sessions will be used mostly to review
and clarify the programming assignments, introduce supplement materials
of the lectures, and answer questions from students.
Course Grades and Assignment Policy
The final grade is based on four programming assignments, one mid-term
exam, and one final exam. Each assignment is 15%, the midterm exam is
20%, and the final exam is 20%. For each assignment, there are also
optional bonus points (up to 20% of its total grade).
All assignments are due on 11:59pm (HK Time) on the due date. In total
there are 4 late days to handle unexpected circumstances (e.g.,
sickness, personal crisis, family problems). If you use up the 4 late
days, we will allow late submissions for up to 24 hours, with a 20%
point penalty. No late submissions after 24 hours are allowed, and zero
marks will be given in that case. All the assignments will be submitted
via Blackboard. More detailed instructions for submission will be
provided along with each assignment.
Students are welcome to discuss their partial solutions and questions
with course staff members during the tutorial sessions, in office hours,
or via Blackboard forum. Students are also permitted to discuss common
concerns with classmates, but these discussions must be kept at a
general level, without exposing their solutions or source code.
Finally, please do not publish any questions or solutions of the
assignments and the exams, e.g., Github or a publicly accessible web
page. This is a violation of the basic Rights, Rules, Responsibilities
of members of the University community.
Academic Honesty
Plagiarism, including copying any parts of code or solutions from your
classmates, or releasing your code for others to copy, or copying during
the exams, is strictly prohibited and will be treated very seriously. If
found responsible, the typical penalty is an F as a course grade plus
whatever penalty that the university imposes. If you have any doubts,
please read the academic honesty guidelines from the university(https://www.cuhk.edu.hk/policy/academichonesty/)
and the academic honesty guidelines from the Faulty of Engineering
(https://www.erg.cuhk.edu.hk/erg/sites/default/files/Guidelines_to_Academic_Honesty.pdf),
or ask the instructor.
Acknowledgement
Some of the materials in the lecture slides of this course is built based on similar courses taught by the following professors:
Prof. Shree Nayar (Columbia)
Prof. Matthew O'Toole (CMU)
Prof. Noah Snavely (Cornell)