Computer Vision - CSCI-GA.2271-001

Semester: Fall 2016.

Time and Location: Thursday 7:10-9:00pm, Warren Weaver Hall, Room 201.

Instructor: Rob Fergus     

Office hours: Thursday 6:10-7:10pm, Room 1226, 12th floor, 719 Broadway.

Grader: TBD

Overview

Computer Vision aims to extract descriptions of the world from pictures or video. In recent years, much progress has been made on this challenging problem. The course will start by looking the established area of a geometric vision. It will then introduce the convolutional network model and describe the profound impact that it has had on problems in recognition, segmentation and many other vision problems.

Prerequisites

The course will be suitable for master's students and advanced undergraduates. A reasonable knowledge of linear algebra will be required, along with some basic concepts in machine learning. The homeworks will require Matlab and Torch, so familiarity with it is desirable, although not essential.

Assessment

Assessment will be through four graded homework assignments.

Late Policy

The policy regarding late homework is as follows: (a) assignments that are late by less than 24hrs will suffer a 10% reduction; (b) those between 24 and 72 hrs late will suffer a 25%x reduction and (c) those more 72hrs late will suffer a 50% reduction. You are strongly encouraged to start the assignment early and don't be afraid to ask for help from either the TA or myself.

Schedule

Date Time Topics Relevant Book Chapters
WEEK 1
Thur 09/08/2016
19:10-21:00 1. Introduction, Image Formation Pt. 1 Szeliski, Ch. 1 and 2; F & P, Ch. 1
Thur 09/08/2016
19:10-21:00 2. Image Formation Pt. 2 (Slides - PPT) (Slides - PDF)
WEEK 2
Thur 09/15/2016
19:10-21:00 3. Filtering & Edges (Slides - PPT) (Slides - PDF) Szeliski, Ch. 3 and 4; F & P, Ch. 6, 7 and 8
Thur 09/15/2016
19:10-21:00 4. Lighting, Color (Slides - PPT) (Slides - PDF)
WEEK 3
Thur 09/22/2016
19:10-21:00 5. Corner & Region detection. Szeliski, Ch. 3 and 4; F and P, ch. 3 and 16; Lowe 2004
Thur 09/22/2016
19:10-21:00 6. Region representation. (Slides - PPT) (Slides - PDF)

Assignment 1 Out (PDF) (assignment1.zip)
WEEK 4
Thur 09/29/2016
19:10-21:00 7. Fitting, RANSAC (Slides - PPT) (Slides - PDF) Szeliski, Ch. 6; F & P sec. 3.1, ch. 15; Winder and Brown 2007
Thur 09/29/2016
19:10-21:00 8. Image Alignment, Optical Flow
WEEK 5
Thur 10/06/2016
19:10-21:00 9. Epipolar geometry Szeliski, Ch. 7; H and Z, ch. 9-12; F and P, ch. 10 and 11
Thur 10/06/2016
19:10-21:00 10. Stereo reconstruction (Slides - PPT) (Slides - PDF) Szeliski, Ch. 7; H and Z, ch. 9-12; F and P, ch. 10 and 11
WEEK 6
Thur 10/13/2016
19:10-21:00 11. Multiview Stereo, Structure from Motion (Slides - PPT) (Slides - PDF)
Thur 10/13/2016
11:00 Assignment 1 Due

Assignment 2 Out (PDF) (world.txt) (image.txt) (sfm_point.mat) (part4.zip)
Thur 10/13/2016
19:10-21:00 12. Structure from Motion
WEEK 7
Thur 10/20/2016
19:10-21:00 13. Introduction to Recognition. (Slides - PPT) (Slides - PDF)
Thur 10/20/2016
19:10-21:00 14. Specific Object Recognition (Slides - PPT) (Slides - PDF) Szeliski, Ch. 14.
WEEK 8
Thur 10/27/2016
19:10-21:00 15. Faces
Thur 10/27/2016
19:10-21:00 16. Recognition - Bag of words models Pt. 1 (Slides - PPT) (Slides - PDF) Szeliski, Ch. 14.
WEEK 10
Thur 11/03/2016
19:10-21:00 17. Recognition - Bag of words models Pt. 2
Thur 11/03/2016
11:00 Assignment 2 Due

Assignment 3 Out (PDF) (faces.zip) (qu2_data.zip)
Thur 11/03/2016
19:10-21:00 18. Recognition - Parts-based models Pt.1 (Slides - PPT) (Slides - PDF)
WEEK 11
Thur 11/10/2016
19:10-21:00 19. Recognition - Parts-based models Pt.2
Thur 11/10/2016
19:10-21:00 20. Neural Networks Part 1 (Slides - PDF)
WEEK 12
Thur 11/17/2016
19:10-21:00 21. Parts and Structure Models Pt. 3 Szeliski, Ch. 5
Thur 11/17/2016
19:10-21:00 22. Segmentation (Slides - PPT) (Slides - PDF)
WEEK 13
Thanksgiving Break (No class)
WEEK 14
Thur 12/01/2016
19:10-21:00 23. Recognition - Boosting (Slides - PPT) (Slides - PDF)
Thur 12/01/2016
11:00 Assignment 3 Due

Assignment 4 Out (PDF)(hw4_qu1.zip) (example_mean_shift.m) (image1.jpg) (tiger.jpg) (baseball.jpg) (Shi and Malik, PAMI 2000)
Thur 12/01/2016
19:10-21:00 24. Recognition - Neural Nets Pt. 2
WEEK 14
Thur 12/08/2016
19:10-21:00 25. Recognition - Neural Nets Pt. 3
Thur 12/08/2016
19:10-21:00 26. Recognition - Neural Nets Pt. 4
WEEK 15
Thur 12/15/2016
19:10-21:00 28. Deep Learning Pt. 1
Thur 12/15/2016
19:10-21:00 27. Deep Learning Pt. 2
EXAM WEEK
Thur 12/17/2016
11:00 Assignment 4 Due (Note that this is a strict deadline)

Acknowledgments

The instructor would like to thank Andrew Zisserman and Svetlana Lazebnik for making their slides available. Thanks also go to Fei-Fei Li and Antonio Torralba for creating the ICCV'05/CVPR'07 object recognition tutorial slides used in classes 11,12,13.

Textbook

The main text book that we will use is:

Szeliski, Richard, Computer Vision: Algorithms and Applications Springer, 2011. This book is available in electronic form at: Link

There are also a couple of other text books relevant to the course, although we won't be directly using them:

Forsyth, David A., and Ponce, J. Computer Vision: A Modern Approach, Prentice Hall, 2003.

Hartley, R. and Zisserman, A. Multiple View Geometry in Computer Vision, Academic Press, 2002.

Both these are available from the CIMS library.

For the object recognition part of the course, please see the Object Reconition Short Course. Link

Additional Material

Matlab guides

Matlab tutorial by Hany Farid and Eero Simoncelli Link

A more comprehensive Matlab tutorial by David Griffiths Link

Further documentation on Matlab can be found here Link

Books

Palmer, Stephen E. Vision Science: Photos to Phenomenology, MIT Press, 1999.

Strang, Gilbert. Linear Algebra and Its Applications 2/e, Academic Press, 1980.

Wandell, Brian A. Foundations of Vision, Sinauer, 1995.