Instructor: Jane Mulligan

CPSC 7000: Topics in Machine Vision is an introductory graduate level course on the use of digital images as a sensor, to extract useful information about the world. A more complete definition of Computer Vision by Pascal Fua.

Prerequisites: C/C++ programming, linear algebra, basic trigonometry, and probability.

syllabus.pdf

Text: Computer Vision -- A modern approach -- David Forsyth and Jean Ponce (available only online at http://www.cs.berkeley.edu/~daf/book3chaps.html)

Topics :

• Image Formation:
• Calibration:
  • Strong
• 2D Vision
  • Filters
  • Binary Images
  • Features
  • Edge Detection
  • Texture, segmentation
• 3D Vision
  • Shape from X
  • stereo
• Dynamic Sequences
  • Optical Flow
  • Structure from Motion
  • Tracking
    • Kalman Filter based
    • Condensation
• Recognition
  • Representations
  • Appearance Based
• What's New

Grading:

• Assignments: 40%
• Project: 60%

Notes etc

• Camera Model: See text Chs. 1 and 6. Lecture notes, Toulouse de Margerie (McGill)
• Radiometry: See text Ch. 2. Lecture Notes , Jepson's Notes (Toronto)
• Color: See text Ch. 4. Lecture Notes
• Calibration: See text Ch. 6. Lecture notes, CV Online Camera Calibration , Autocalibration
• Linear filters: See Ch. 8. Lecture Notes .
• Fourier, noise etc: See Ch. 8. Lecture Notes .
• Binary Images: See morphology Ch. 10. Lecture Notes .
• Image Features (edges): See Ch 9. Lecture Notes .
• Corners, Hough Transform: See Ch 10.3 and Ch. 17. Lecture Notes .
• Ellipses, Deformable Contours, Ch. 17. Lecture Notes .
• Texture, Ch. 11. Lecture Notes .
• Segmentation and Clustering, Ch. 16. Lecture Notes .
• Probabilistic Segmentation, Ch. 18. Lecture Notes .
• Tracking, Ch. 19. Lecture Notes .
• Particle Filtering, Ch. 19. Lecture Notes .
• Motion Field. Lecture Notes ; CVOnline.

Assignments

• Assignment 1: Calibration ; pdf
• Assignment 2: Background Subtraction.
• Assignment 3: Feature extraction and tracking.

Project: Instructions.