Fundamentals of Computer Vision     COMP 558
Fall 2018 
  Location and Time TBD

Instructor:    Professor Michael Langer
Office:        McConnell Engineering, rm. 329
Tel:             514-398-3740
Office Hours:           TBD
Teaching Assistants (T.A.):

Office :   
Office Hours:                     (by appointment)
  • I will co-teach the course with Prof. Kaleem Siddiqi in Fall 2018.
    Below is the approximate lecture schedule and topics.
  • Here is a link to my 2010 version.
  • Minerva says that COMP 360 is a prerequisite but that is overkill. In terms of
    algorithms, COMP 250 is enough. The more important prequisites are the
    MATH courses. You need to be very comfortable with multivariable Calculus
    (MATH 222) and linear algebra (MATH 223).
  • The Assignments will use Matlab which is available for free to McGill students
Resources and other links

Lecture Schedule


  1. Course Outline [MK, TBD]

2D Vision: Image processing and representation

  1. RGB [M]
  2. Image filtering [M]
  3. Edge detection [M]
  4. Lines and vanishing point detection (Hough and RANSAC) [M]
  5. Scale space (Gaussian) [K]
  6. Features 1: corners, intro to histograms [K]
  7. Features 2: histogram-based (SIFT, HOG) [K]
  8. Features 3: learned features (CNN's) [K]
  9. Image Registration (Lucas-Kanade) [K]
  10. Tracking (KLT + Shi/Tomasi) [K]
  11. Segmentation and Tracking (Mean Shift) [M]


3D Vision

  1. Linear perspective, camera translation [M]
  2. Vanishing points, camera rotation [M]
  3. Homogeneous coordinates, camera intrinsics [M]
  4. Least Squares methods (eigenspaces, SVD) [M]
  5. Camera Calibration + intro homographies [K]
  6. Homographies & rectification [K]
  7. Stereo and Epipolar Geometry [K]
  8. Stereo correspondence [M]
  9. Appearance models 1: lighting and reflectance [M]
  10. Appearance models 2: photography [M]
  11. RGBD Cameras & Point Clouds 1 [K]
  12. RGBD Cameras & Point Clouds 2 [K]
  13. Guest lecture (Dave Meger)