COMP 498G/691G Computer Vision - Charalambos Poullis
 

COURSE SCHEDULE

The table below outlines a tentative schedule for this course over a 13-week term. In conjunction to the lectures there will be a 2-hour weekly lab session intended to give practical demonstration of the computer vision principles presented in the lectures and to provide experience in using OpenCV for the development of computer vision components and systems.

Ackowledgements: The slides are a combination of multiple resources and materials generously made publicly available by L. Shapiro, J. Hays, S. Lazebnik, D. Forsyth, J. Ponce, J. Koenderink, S. Seitz, R. Szeliski, B. Freeman, M. Pollefeys, D. Lowe, K. Grauman, A. Efros, F. Durand, L. Fei-Fei, A. Torralba, R. Fergus, F-F. Li, A. Karpathy, J. Johnson. In particular the material is heavily based on Professors' L. Shapiro's and J. Hays' slides.

2018 Winter Semester
Date Topics Chapters Slides Comments
1 > Syllabus
Introduction to Computer Vision
Images
Szeliski Ch. 1, 3.2, Forsyth/Ponce Ch. 4 pdf
pdf
pdf
Assignment 1 out
2 > Image sampling
Edge detection
Szeliski Ch. 3.2, 3.4, 3.5, 4.2 pdf
pdf
Cipolla & Gee on edge detection
ink
ink
3 > Geometric transformations
Interest points
Szeliski 4.1.1 pdf
pdf
Assignment 1 due
Assignment 2 out
Harris paper
ink
ink
4 > Feature Descriptors
Image Stitching I
Szeliski 4.1.2-4.1.3
Szeliski 6.1
pdf
pdf
SIFT paper
ink
ink
5 > Quiz #1
Assignment 1 Solution
6 > Image Stitching II
Cameras
Szeliski 6
Szeliski 2
pdf
pdf
ink
ink
Assignment 2 due
7 > Multiple Views (stereo, epipolar geometry) Szeliski Ch. 9 pdf
pdf
8 > Multiple views and motion (structure from motion, Motion and Optical Flow Szeliski Ch. 7
Szeliski Ch. 8.4
pdf
pdf
Project out
9 > Sliding Window Face Detection with Viola-Jones
Assignment 2 Solution
Szeliski Ch.14.1, 14.2 pdf
10 > Image Classification
Loss Function and Optimization
pdf
pdf
11 > Back-propagation and Neural Networks
Training Neural Networks
pdf
pdf
12 > Quiz #2
13 > Convolutional Neural Networks pdf