This course is an introductory graduate course in computer vision. We will cover principles of image formation, local feature analysis, multi-view geometry, image warping and stitching, structure from motion, and visual recognition. We will also touch upon related topics in signal and image processing including convolution and image pyramids, and may cover computer graphics topics involving computational photography and image-based rendering as time permits.
This course is appropriate as a first course for graduate students with an EECS background, which should have prepared the students with these essential prerequisites:
· Data structures
· A good working knowledge of MATLAB programming (or willingness and time to pick it up quickly!)
· Linear algebra
· Vector calculus
The course does not assume prior imaging experience, computer vision, image processing, or graphics.
There will be three equal components to the course grade
· Five problem sets
· A take-home exam, tentatively scheduled for Nov 11-13.
· Final project (including evaluation of proposal document, in-class presentation, and final report)
In addition, strong class participation can offset negative performance in any one of the above components.
You may discuss the problem sets with others, but you must submit individual work. No discussion or collaboration is allowed on the take-home. The final project should include a significant novel implementation of a technique related to the course material. Alternatively, a journal length review article is acceptable. Teams of 2 encouraged for implementation projects (but document role throughout!). The project proposal should be no more than 5 pages; the in-class presentation should be 10 minutes, and the final report should be no longer than 15 pages.
The course has a website on https://bspace.berkeley.edu/, I will use that to post lecture notes, papers to read, homework assignments etc. If you are registered for the course, you have automatic access. In special cases, I can add people as “guests”. CLICK HERE FOR A LECTURE NOTE MIRROR SITE.
The primary course text will be Rick Szeliski’s draft Computer Vision: Algorithms and Applications; we will use an online copy of the June 19th draft. A copy and link will be provided in bSpace. The secondary text is Forsyth and Ponce, Computer Vision: A Modern Approach, and is available at the bookstore and amazon.com, etc.
Problem sets and projects will involve Matlab programming (you are free to use alternative packages). Matlab runs on all the Instructional Windows and UNIX systems. Instructions and toolkits are described in http://inst.eecs.berkeley.edu/cgi-bin/pub.cgi?file=matlab.help. CS280 students can use their existing EECS Windows accounts in EECS instructional labs, and they can request new accounts (for non-majors) or additional access to Instructional resources by following the instructions about ’named’ accounts in http://inst.eecs.berkeley.edu/connecting.html#accounts. They can logon remotely and run it on some of our servers: http://inst.eecs.berkeley.edu/connecting.html#labs
The course schedule will be maintained as a Google calendar “UCBC280CAL” available at: http://tinyurl.com/UCBC280CAL and which will also be linked directly from within the bSpace site.
ASSIGNED READINGS WILL BE MARKED IN THE CALENDAR ON THE ASSIGNED DATE WITH THE AUTHOR AND SECTION, e.g. “Szel., 2.1-2.3” for Chapter 2.1-2.3 in the primary text. CHECK THE ONLINE CALENDAR WEEKLY FOR UPDATES.
Problem set assignment will be distributed via the assignments page in bSpace. Course project documents and presentations should be uploaded via bSpace. Deadlines for both will be listed in the online calendar.
Thursday 5pm-6pm (except for 9/10) in Soda 413, or by appointment.