CLASS INFORMATION:

Course Description: -- Complete list of Topics.

Today's computers interact in a limited  way with the world and with humans because they lack the ability to "see".  Cameras and video recorders capture visual information without understanding the information they capture.  The goal of computer vision is to "discover from images what is present in the world,  where things are located, what actions are taking place" (Marr 1982).  To achieve this goal, we need to know how light is reflected off surfaces, how objects move, and how this information is projected onto an image by the optics of a camera.  Unfortunately, information is lost when the three dimensional world is projected onto a two dimensional image. We will study the mathematics needed to devise algorithms to recover, or reconstruct, some of physical properties of the world from one or more images.  However,  vision is more than simply reconstructing the 3D world from 2D images, it is about image "understanding".

This course is an introduction to basic concepts in computer vision, as well some research topics. We will cover low-level image analysis, image formation, edge detection, segmentation, image transformations for image synthesis, methods for 3D scene reconstruction, motion analysis, tracking, and object recognition.

Prerequisites: Linear Algebra, Probability, or consent of the instructor.

Grading:

			Option A	Option B
	Problem Sets (~6) with lab exercises in Matlab. Problem sets may be discussed, but all written work and coding must be done individually.		40%	60%
	Exams: One take-home exams.              (Take-home exam may not be discussed.)		20%	0%
	Final Project:   An original implementation of a new or published idea A detailed empirical evaluation of an existing implementation of one or more methods   Project proposal not longer than two pages must be submitted and approved before the end of March.		30%	30%
	Class Participation:		10%	10%

Late Policy:  Late assignments will not be accepted without prior approval. Get prior approval. No exceptions!