Overview

This course will cover the essentials of computer vision. It is a graduate-level course of interest to anyone seeking to process images or camera information, or to acquire a general background in issues related to real-world perception and computational geometry.

Activities

The course involves three types of activities:

• Interactive classroom sessions, where students together with the instructor explore the basic mathematical foundations behind a range of popular algorithms.

• Homework assignments will provide an opportunity to deepen the problem solving skills acquired in class.

• This year we are trying something new: a software research project. We, the instructors, will provide video data acquired by a moving platform. The research project will require students to develop software for extracting information from this video stream, building on the many algorithms discussed in class. The instructors will evaluate the software and reward the authors of the systems with the best performance.

Prerequisites

This is an introductory graduate level course. Familiarity with basic statistical concepts (Bayes rule, PDFs, projective geometry, Kalman filters, continuous distributions...) will be helpful for this course, as will be hands-on experience with software development in C or C++ and Matlab. Intro tutorials will be given into Matlab and the vision library OpenCV. But the most important prerequisite will be creativity and enthusiasm.