About this courseHumans are extremely good at perceiving the world from visual input alone. This comes so easily to us that we underestimate how difficult perception it is, and how hard it is for machines, as the webcomic above illustrates.
Computer vision is a subfield of AI focussed on getting machines to see as humans do. We have been at it for almost half a century now, and while the problem is still far from getting solved, we have made tremendous progress. The past decade has especially been a revolution in the making. This course will cover these ideas, both the classic work based on geometry and physics, as well as the new ones based on convolutional networks and deep learning.
Quick infoInstructor: Bharath Hariharan
Lecture time: Tues. and Thurs. 1:25pm - 2:40pm
Lecture venue: Gates G01
TA: Guandao Yang
Bharath: Tue and Thur 3-4 pm (at 311 Gates Hall)
Guandao: Fri 4-5 pm Gates G17
Intended audience and prerequisites:This course is intended for PhD students. As such, it will assume familiarity and comfort with mathematics in general and linear algebra and probability and statistics in particular. This intended audience also means that there will be a strong focus on teaching research abilities (see below for what this means).
If you are a PhD student and you find the course full, please send me a note and I will help you enroll. If you are an undergraduate student interested in this course, I strongly recommend you take the undergraduate version (CS4670/5670) that is offered in the spring. I will not prevent you from enrolling. However, if the course is full, I will not be able to help you.
|Learning outcomes||Calendar and lecture notes||Resources||Grading policy|