CS4670/5670 - Introduction to Computer Vision

Picture credit: Magritte and some computer vision researchers

Quick info

Instructor: Bharath Hariharan
Lecture time: Mon. / Wed. / Fri. 1:25pm - 2:10pm
Lecture venue: Phillips Hall 101
Piazza
TAs:
  • Jimmy Briggs
  • Athena Huang
  • Zhiqiu Lin
  • Yiwei Ni
  • Karun Singh
  • Alvin Zhu
Office Hours:
  • Bharath: M/W/Thur 10-11 am (at 311 Gates Hall)
  • Jimmy Briggs: Tu 10-11 am (Gates G13) and Fri 10-11 am (Gates G11)
  • Athena Huang: Tu 7-9pm (Gates G11)
  • Zhiqiu Lin:
  • Yiwei Ni: W 5-7 pm (Gates G15)
  • Karun Singh: Fri 2:30-4:30 pm (Rhodes 405)
  • Alvin Zhu: W 3:30-5:30 (Gates 406)

Overview:

This course will serve as a detailed introduction to computer vision. The emphasis will be on covering the fundamentals which underly both computer vision research and applications. A tentative list of topics is below: We will have a combination of about 4 programming assignments and 2 written homeworks. This is subject to change.

Office Hour Calendar

Lectures / Notes:

Reference: Rick Szeliski's book. This is not a textbook, in that it covers a lot more material in a lot more detail, but can be used for additional reading.
Course plan: Below is the (tentative) list of classes, with possible additional readings. These may change as the semester progresses.
Date Topic (with linked notes / slides) Additional reading Assignments etc
Jan 24 Introduction [ppt | pdf] Szeliski 1 -
Jan 26 The visual world [ppt | pdf] Szeliski 2 -
Jan 29 Image filtering [ppt | pdf] Szeliski 3.1-3.2 -
Jan 31 Image filtering and Fourier transforms [ppt | pdf] Szeliski 3.4 -
Feb 2 Fourier transforms and resizing and resampling [ppt | pdf] Szeliski 3.4, 2.3.1 -
Feb 5 Resizing, resampling and pyramids [ppt | pdf] Szeliski 3.5, 2.3.1 -
Feb 7 Grouping I - Edge detection [ppt | pdf] Szeliski 3.5, 4.2 -
Feb 9 Numpy / scipy tutorial [ppt | pdf] - -
Feb 12 Grouping II - Edge detection and k-means [ppt | pdf] Szeliski 4.1, 5.3 -
Feb 14 Grouping III - Images as graphs[ppt | pdf] Szeliski 5.3 PA1 due
Feb 16 Grouping IV | The correspondence problem [ppt | pdf] - -
Feb 19 February break - -
Feb 21 Feature detection [ppt | pdf] Szeliski 4.1 -
Feb 23 Harris corner detector [ppt | pdf] Szeliski 4.1 -
Feb 26 Feature descriptors and matching - I [ppt | pdf] Szeliski 4.1 -
Feb 28 Feature descriptors and matching - II [ppt | pdf] Szeliski 4.1 -
Mar 2 --Snow day-- Szeliski 2.1 -
Mar 5 Feature descriptors and matching - III | Geometry of image formation - I [ppt | pdf] Szeliski 2.1 -
Mar 7 Geometry of image formation - II [ppt | pdf] Szeliski 2.1 -
Mar 9 Homogenous coordinates | Camera calibration - I [ppt | pdf] Szeliski 2.1, 6.1, 6.2 -
Mar 12 Prelim review - -
Mar 14 Camera calibration - II | Triangulation [ppt | pdf] - -
Mar 16 Homographies | RANSAC [ppt | pdf] Szeliski 6.1 -
Mar 19 RANSAC and Hough transforms [ppt | pdf] Szeliski 6.1 PA3 out
Mar 21 Prelim Discussion | Two-view stereo - I [ppt | pdf] Szeliski 7.2 -
Mar 23 Two-view stereo [ppt | pdf] Szeliski 7.2 -
Mar 26 Epipolar geometry [ppt | pdf] Szeliski 7.1-7.4 -
Mar 28 Epipolar geometry - II [ppt | pdf] Szeliski 7.1-7.4 -
Mar 30 Radiometry [ppt | pdf] Szeliski 2.2 PA3 due
Apr 2 Spring break - -
Apr 4 Spring break - -
Apr 6 Spring break - -
Apr 9 Photometric stereo - I [ppt | pdf] Szeliski 12.1.1 -
Apr 11 Photometric stereo - II [ppt | pdf] Szeliski 12.1.1
Practice questions for stereo
-
Apr 13 Photometric stereo - III | Intro to recognition [ppt | pdf] - -
Apr 16 Intro to machine learning - optimization | the ERM principle [ppt | pdf] - -
Apr 18 Machine learning and optimization | the ERM principle [ppt | pdf] Answers for stereo practice questions
Practice questions for photometric stereo
-
Apr 20 Regularization | Linear classifiers and HOG / SIFT Bag-of-words [ppt | pdf] - -
Apr 23 Linear classifiers and HOG/SIFT - -
Apr 25 Bag-of-words and non-linear classifiers - -
Apr 27 Neural networks and backpropagation - -
Apr 30 Convolutional networks - -
May 2 Image classification and transfer learning - -
May 4 Object detection - -
May 7 Semantic segmentation - -
May 9 Conclusion - -