Computer Vision (CS570) - Department of Computer Science

Course Title: Computer Vision
Course Code: CS570
Credits: 3
Level: Graduate
Instructor: Dr. Nazar Khan
Semester: Fall, 2024
Location: FCIT, Allama Iqbal Campus, Alkhwarizmi Lecture Theater (AKLT)
Class Times: Mondays and Wednesdays, 12:15 AM - 2:15 AM 		Textbook:
Foundations of Computer Vision (FoCV)
by Antonio Torralba, Phillip Isola, and William T. Freeman.
MIT Press
April, 2024
Join Google Classroom 		FoCVcover

Welcome to the Computer Vision course page! Computer Vision is a field of AI that focuses on enabling machines to interpret and understand the visual world. In this course, you will study both the theoretical foundations and practical applications of computer vision techniques, exploring how algorithms can process, analyze, and extract meaningful information from images. This course covers the theory and practice of computer vision. Dive into image processing, neural networks, object recognition, and advanced techniques such as deep learning, 3D vision, and motion analysis.

Lectures

Lecture Topic Readings Tutorials & Evaluations
1 Introduction Ch 1 Setup a Python environment for CV
2 Looking at Images + Computer Vision and Society Ch 3, 4
3 Image Formation and Lenses Ch 5, 6 How to make a pinhole camera
Tutorial 1: Introduction
4 Cameras as Linear Systems + Color Ch 7, 8
  • See first 8 slides of this lecture on matrix and vector calculus. You may also see the video explanation of those slides.
  • See the Wiki page on matrix calculus from which we will use denominator-layout notation.
  • See pages xxvii, xxviii, xxx, and xxxi of the FoCV textbook.
 Quiz 1, Solution
Tutorial 2: Color Spaces
5 Introduction to Machine Learning Ch 9 Slides Tutorial 3: Linear Regression
6 Gradient-based Learning + Generalization Ch 10, 11 Slides, Slides
7 Neural Networks Ch 12, 13 Slides Tutorial 4: Neural Networks using PyTorch: Recognition of Handwritten Digits
8 Linear Image Filtering Ch 15 Quiz 2, Solution
Tutorial 5: Convolution
9 Blur Filters Ch 17
10 Image Derivatives Ch 18 Assignment 1
Deadline: 11:59 pm, November 27
11 Image Sampling Ch 20, 21
12 Image Pyramids Ch 23 Assignment 2
13 Convolutional Neural Networks Ch 24 Quiz 3, Solution
14 Transformers Ch 26 Assignment 2
Deadline: 11:59 pm, January 5
15 Representing Images and Geometry Spatial Transformations Quiz 4, Solution
16 Estimating and Applying Spatial Transformations Estimating and Applying Spatial Transformations
17 RANSAC Robust Estimation via RANSAC Quiz 5, Solution
18 Camera Modelling Camera Geometry
19 Camera Anatomy Camera Anatomy
20 Camera Calibration Camera Calibration
21 Epipolar Geometry Epipolar Geometry
22 Stereo Reconstruction Stereo Reconstruction

Grading

Assignments Project Quizzes Midterm Exam Final Exam
Weight 12% 8% 5% 35% 40%