About Us
Faculty Profiles
Faculty Advisers
Placements Statistics
Activity Chart
Course Coordinators
Contact us
Computer Vision

Welcome to the webpage of MAN-522 Computer Vision course.


How can computers understand the visual world of humans? This course treats vision as a process of inference from noisy and uncertain data and emphasizes probabilistic, statistical, data-driven approaches. The course will start with some mathematical prelimnaries and then topics like image formation and processing; camera calibration, stereo vision, edge and boundary detection; and structure from motion. This offering of MAN-522 will emphasize the core vision tasks of scene understanding and recognition. We will try to cover few classification case studies in order to create an interests among student for going to higher studies/industrial research in this area.


  • Computer Vision: Algorithms and Applications, Richard Szeliski, ISBN-10: 1848829345 (online version available)
  • Computer Vision: A Modern Approach (2nd Edition), David A. Forsyth and Jean Ponce, Prentice Hall, ISBN-10: 013608592X

Reading Materials and Notes:

  1. Mathematical Preliminaries: SVD and Least Square
  2. Image Processing: Formation and Fundamentals (Part-I and Part-II)
  3. Pinhole Camera and Projections: Camera Parameters
  4. Camera Calibration: Indirect and direct methods
  5. Stereo Vision: Fundamental and Essential matrices
  6. Features Extraction part-1: edges, corners, and blobs
  7. Features Extraction part-2: Hough transform and model fitting
  8. Object Recognition part-1: PCA
  9. Object Recognition: Part-2: LDA
  10. Face Recognition
  11. Optical Flow
  12. Shape from Shading
  13. Structure from Motion
  14. Linear Classifiers



Assignment 1 to be posted soon