ls

Lecture Topics
  1. Tuesday, Jan 15. Introduction to Class. No Textbook. Encouragement to use matlab. All vision is hard. Computer Vision is hard; very interdisciplinary (statistics, geometry, optimization, probability, linear algebra, some psychology). Computer Vision is different than image (or video) processing. It is also different than photogrammetry.
    • Lecture Notes (ppt)
    • General resource on face detection: (link)
    • Reading: PAMI 2002 paper. Read up to Section 2.2 and then Section 3.3.
    • Assignment 1: Due, January 22. In the language of your choice, create a program that reads in an image, finds the eyes, and outputs the image with the eye location marked. Turn in:
      (a) a pseudo-code description of your eye-detection algorithm. This description should have no more than 100 words. (b)examples of your program working on at least 1 image, and an example where your program fails.
  2. Thursday, January 17. Faces are compelling natural objects. Template matching as applied to face detection, introducing SSD template match and correlation based template match.
    • Lecture Notes (ppt)
    • Download the matlab code from today's class here, and, if you can't find a picture on your own, the picture we used today here.
    • Description of the KISMET project, discussed in class link
    • The strange facts of prosopagnosia (i especially like the story about stones), and a rather bizarre movie about it, html-link.
  3. Tuesday, January 22. Faster face detection. Using AdaBoost of simple image features.
  4. Thursday, January 24. PCA on face images
  5. Tuesday, January 29. PCA reprise, and intro to camera geometry
  6. Thursday, January 31. Camera Calibration, and some linear algebra
  7. Tuesday, Febuary 5. Camera Coordinate Transforms, Homographies
  8. Thursday, Febuary 7. Stereo, Fundamental Matrix, Essential Matrix
  9. Tuesday, Febuary 12. Stereo-Reprise, Rectification, point matching.
  10. Thursday, Febuary 14. Motion Fields. Continuous Camera Motions, Optic Flow.
  11. Project 2 details
  12. Tuesday, Febuary 19. Computing Optic Flow.
  13. Thursday, Febuary 21. Motion Segmentation.
  14. Tuesday, Febuary 26. Background Modeling
  15. Thursday, Febuary 28. Dynamic PCA
  16. Image features
  17. Shape Contexts
  18. SIFT
  19. Object Recognition, Bag of Words
  20. Project 3 initial page.
  21. Intro to segmentation and spectral methods
  22. Intro to segmentation by clustering
  23. Intro to tracking
  24. Intelligent scissors and snakes
  25. Level Sets
  26. Lecture 26
  27. Lecture 27, Shape from Shading
  28. Lecture 28, scattering

    Final Project details are available at: link.