E1 216: Computer Vision

2024 Edition

Course Webpage

Course Outline :

Of all the human senses, vision is the richest in content and perhaps the hardest to formalise in a rigorous manner. As a discipline, Computer Vision covers a wide variety of methods for interpretation and analysis of visual data using a computer. In this course we will present a broad, introductory survey of the field. The objective of the course is to develop a familiarity with the approaches to modelling and solving problems in computer vision. Mathematical modelling and algorithmic solutions for vision tasks will be emphasised.

We shall endeavour to cover the following topics:

• Camera Geometry
• Radiometry and Applications
• Image Features
• Robustness
• Geometric Transformations
• Image Segmentation
• Object Detection + Recognition
• Learning in Computer Vision
• Stereo and Depth Cameras
• Camera Calibration
• Geometry of Two Cameras
• The Shape of the World: Large-scale 3D Reconstruction
• Optical Flow
• Ethics in Computer Vision

Course Textbooks

There is no prescribed textbook for the course. We will use material from these books and a number of other sources.

1. Computer Vision : A Modern Approach by David Forsyth and Jean Ponce, Pearson India, 2015.
   We will use some chapters from this book.
2. Computer Vision : Algorithms and Applications by Richard Szeliski. Draft 2nd Edition.
3. Computer Vision: Models, Learning, and Inference by Simon J D Prince, 2012.
   Interesting volume using a unified language of probability. Some lectures will draw from this book.
4. Multiple View Geometry in Computer Vision by R. Hartley and A. Zisserman, Second Edition, Cambridge University Press, 2004.
   We will use the sample chapters available online.