Introduction to Computer Vision   

 

Spring 2020

Basic Information:

·  Lecture time: Mon/Wed 5:30pm – 6:50pm, Engineering 143, West Campus

·  Instructor: Prof. Haibin Ling, hling AT cs.stonybrook.edu

·  Office hour: by appointment

·  Syllabus: PDF

·  TA: Tao Sun, Tao.Sun.1 AT stonybrook.edu,

                Office hour: Thur 3-5pm, by appointment.

·  TA: Kunal Kolhe, kkolhe AT cs.stonybrook.edu

                Office hour: Tues 10-11am, by appointment.

 

Useful Links:

·  Computer Vision: Algorithms and Applications, Richard Szeliski, 2010.

·  Computer Vision:  Models, Learning, and Inference, Simon J.D. Prince, 2012.

Date

Topics and Schedule (up to change)

Material

Week 1
1/27

General introduction

·       Background

·       Topics and applications

·       Related fields

·       Milestones

 

1/29

Image formation

·       Basic facts about light

·       Anatomy of cameras

·       Matting

Szeliski 2.1, esp. 2.1.5

Week 2
2/3

Filters and Noises

·       Modeling image noise

·       Convolution

Szeliski 3.1, 3.2, 3.3.1, 3.4 (Fourier Transforms)

2/5

·       Smoothing images

·       Image pyramids

 

Week 3
2/10

Image Features

·       Point features

·       Corners and Edges

·       Scale

·       Orientation  

Szeliski 4.1.1, 4.1.2, 4.2

2/12

Guest lecture by Prof. Xianfeng David Gu

 

Week 4
2/17

Model Fitting

·       Lines, curves

·       Hough Transform

Szeliski 4.1.3, 4.3.2

Szeliski 6.1, 2.1

2/19

·       Deformation

·       RANSAC

 

Week 5
2/24

Perspective Projection

·       Homogeneous coordinates

·       Image warping, mosaics

Szeliski 2.1

2/26

·       Midterm review

 

Week 6
3/2

Multiple View Geometry

·       Stereo viewing and reconstruction

·       3D range scanning

Szeliski 6.3.1, 7, 11

 

3 /4

·       Midterm 1

 

 

Week 7
3/9

Object Recognition

·       Overview of machine learning in computer vision

·       PCA for image patches

 

3/11

·       Object representation

·       Classifiers

·       Object categories

 

 

Spring Break

 

Week 8
3/30

Deep Learning

·       Convolutional neural networks

·       Architectures

· Deep Learning, Goodfellow et al., 2016, MIT press. http://www.deeplearningbook.org/

4/1

·        Applications

· Deep Learning Tutorial, http://ufldl.stanford.edu/tutorial/.

Week 9
4/6

Deep Learning Practice

·       Pre-training

 

4/8

·       Data-augmentation

 

 

Week 10
4/13

Illumination

·       Shading, shadows

·       Reflectance properties

 

4/15

·       Midterm 2

 

Week 11
4/20

Deep Generative Models

·       Autoencoders, VAEs

·       Generative adversarial networks (GAN)

 

4/22

·       Application in computer vision  

 

Week 12
4/27

Detection

·       Traditional solutions

·       Deep learning solutions

 

4/29

·       Semantic segmentation  

 

Week 13
5/4

Motion & Tracking

·       Low level motion

·       Flow

Szeliski 8.4

5/6

·       Recurrent neural networks

Tracking in 2D and 3D

 

 

 

 

TBD

Final exam