CPSC 425 Computer Vision - Winter 2019/20

Term 1 (September – December, 2019)

Course Description

Computer vision, broadly speaking, is a research field aimed to enable computers to process and interpret visual data, as sighted humans can. It is one of the most exciting areas of research in computing science and among the fastest growing technologies in today’s industry. This course provides an introduction to the fundamental principles and applications of computer vision (see topics).

Prerequisite: All of MATH 200, MATH 221 and either (a) CPSC 221 or (b) all of CPSC 260, EECE 320.

People

Instructor: Jim Little (little [at] cs.ubc.ca).
Office hours: 10-11am Wednesdays

TA: Alex Fan (fan [at] cs.ubc.ca). Office hours: ??/
TA: Rayat Hossain (rayat137 at cs.ubc.ca). Office hours: Thursday 2PM-3PM Room X151
Friday 11AM-Noon Room X151/
TA: Farnoosh Javadi (fjavadi [at] cs.ubc.ca). Office hours: ??/

Topics

Here is a rough schedule and tentative list of topics and readings (subject to change).

Important Dates

First Day of Classes 4 September (Wednesday)
Add/Drop Deadline 17 September (Tuesday)
Drop with W Deadline 11 October (Friday)
Midterm Exam ???
Last Day of Classes 29 November (Friday)
Final Exam December !!!!! but consult exam for official location/time

Lectures

Mondays, Wednesdays and Fridays, 4:00pm-450pm, DMP ???
We will post the lecture materials here.

Sep. 4 Introduction Lecture 1
Sep. 6 Image Formation Lecture 2
Sep. 9 Image Formation Lecture 3
Sep. 11 Image Filtering Lecture 4
Sep. 13 Image Filtering Lecture 5
Sep. 16 Image Filtering Lecture 6
Sep. 18 Sampling slides Reading: Forsyth & Ponce (2nd ed.) 5.2
Sep. 20 Sampling Template Matching slides Reading: Forsyth & Ponce (2nd ed.) 5.3.0-5.3.1
Sep. 23 Scaled Representations slides Reading: Forsyth & Ponce (2nd ed.) 5.3.0-5.3.1
Sep. 25 Local Image Features slides Reading: Forsyth & Ponce (2nd ed.) 5.3.0-5.3.1
Sep. 27 Local Image Features slides Reading: Forsyth & Ponce (2nd ed.) 5.3.0-5.3.1
Sep. 30 Local Image Features Texture slides Reading: Forsyth & Ponce (2nd ed.) 5.3.0-5.3.1
Oct. 2 Local Image Features slides Reading: Forsyth & Ponce (2nd ed.) 6.1,6.3
Oct. 4 Texture slides Reading: Forsyth & Ponce (2nd ed.) 3.1-3.3
Oct. 7 Texture slides
Oct. 9 Colour Model Fitting slides
Oct. 11 Midterm review slides Reading: Forsyth & Ponce (2nd ed.) 5.4
Oct. 16 Midterm slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 18 Local Image Features SIFT slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 20 Optical Flow slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 21 Model Fitting slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 23 Model Fitting slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 25 Model Fitting slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 28 Stereo slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Oct. 30 Stereo slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2 Reading: Transformations: Szeliski pp 38-39
Nov. 1 Optical Flow slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2
Nov. 4 Grouping slides Reading: Forsyth & Ponce (2nd ed.) 5.4, 10.4.2
Nov. 6 Classification slides (cameras) Reading: Forsyth & Ponce (2nd ed.) 1.1.1–1.1.3
Nov. 8 Classification slides Reading: Forsyth & Ponce (2nd ed.) 6.2.2, 9.3.1, 9.3.3, 9.4.2
Nov. 13 Scene Classification slides Reading: Forsyth & Ponce (2nd ed.) 6.2.2, 9.3.1, 9.3.3, 9.4.2
Nov. 15 Classification slides Reading: Forsyth & Ponce (2nd ed.) 6.2.2, 9.3.1, 9.3.3, 9.4.2
Nov. 18 Object Detection slides Reading: Forsyth & Ponce (2nd ed.) 6.2.2, 9.3.1, 9.3.3, 9.4.2
Nov. 20 Object Detection slides Reading: Forsyth & Ponce (2nd ed.) 16.1.3-16.1.4
Nov. 22 Object Detection Grouping slides Reading: Forsyth & Ponce (2nd ed.) 17.1-17.2
Nov. 25 Grouping Grouping NNs slides Reading: Forsyth & Ponce (2nd ed.) 17.1-17.2
Nov. 27 CNNs slides Reading: ConvNet tutorial
Nov. 29 Final Review slides Reading: None

Discussion

Piazza group - Please use this link to enroll yourself.

Assignments

There are seven assignments given throughout the term. The first is a self-study tutorial introduction to Python for computer vision (that is not marked). The other five are handed in to be marked. Each assignment has a specific due date and time which will be announced here (a tentative schedule is posted).

Assignment 0 No due date (but try to complete it by Sep. 11)
Assignment 1 Due 1159pm Tuesday Sep. 24
Assignment 2 Due on Oct. 2
Assignment 3 Cover Sheet Due on Oct. 14
Assignment 4 Cover Sheet Due Wednesday Nov. 14
Assignment 5 Cover Sheet Due Wednesday Nov. 28
Assignment 6 Cover Sheet Due Thursday Apr. 6

Exams

There will be one midterm and one final exam (see important dates).
The midterm is closed-book. The final exam is also closed book.

Grading Scheme

In-class (clicker questions): 5%
Assignments: 25%
Midterm exam: 25%
Final exam: 45%

The instructor reserves the right to change this scheme (but does not anticipate using that right).

Textbook

The course uses the following textbook, which is recommended (but not required):

Another useful textbook (which can be downloaded from http://szeliski.org/Book/) is:

Here is another one which can also be freely downloaded as a PDF from SpringeLink, through UBC Library (must login using CWL).

The following textbooks are also on reserve in the reading room:

Interesting Links

Interesting links

Course Policies