image of UIUC logo

ECE 598 YM - An Invitation to 3-D Vision

Fall 2004, ECE, University of Illinois at Urbana-Champaign

 
"Vision is the art of seeing things invisible." 
                                        
                            -- Johnathan Swift

[Administrative | Homeworks | Handouts | Projects | Online Resource]

Administrative Information

 Instructor: Professor Yi Ma
Lectures: TuTh 2:30am-3:50pm, 245 Everitt Lab
Office hours: W1:30pm-3:00pm, 145 CSL
Office: 145 CSL, Phone: 244-0871
Email: yima(at)uiuc.edu
Appointments: through email.

Announcements:


Description (Full PDF Version):
This is a course for graduate students in ECE, CS, ME or Math interested in building 3-D graphical models from 2-D images, recovering 3-D ego-motion of mobile robots from vision sensors, or solving the inverse problem from 2-D imagery to 3-D geometry in general. More specifically, it will cover the most recent (1-2 years) advances in geometric, algebraic and algorithmic aspects of recovering 3-D motion and structure from a sequence of 2-D images, with a special emphasis on analysis of multiple views and multiple motions. Although it is an advanced topic, the course will provide a self-contained (and comprehensive) coverage on a Multiple View Approach to the recovery of 3-D information from 2-D images: from rigid-body motion, image formation, feature extraction, all the way up to a unified theory/algorithm/system for 3-D reconstruction from multiple views of multiple moving objects.
Prerequisites:
There is no a priori knowledge in computer vision required. This course may be taken independently of or together with the other Computer Vision course CS443/ECE449. The prerequisites are very much the same.
*This course does require a solid background in linear algebra (Math 318, Math381 or ECE415).
*Familiarity with Matlab is recommended . If you have never used it before, it only takes a couple of weeks to learn (say from Matlab Primer).
*Some familiarity with rigid body kinematics (ECE389/GE389), graphics (CS318), image (signal) processing (ECE447), geometry (Math423 or Math424), linear systems theory (ECE415), or estimation theory (ECE461) will certainly increase your appreciation but not crucial nor required.
Required Text:
*An Invitation to 3-D Vision: From Images to Geometric Models, Yi Ma, S. Soatto, J. Kosecka, and S. Sastry, Springer-Verlag, November 2003. If the university bookstore rans out of copies, you may get it at springer.com or amazon.com or barnesandnoble.com. Here is an updated errata of the book.
Supplementary Text:
* Multiple View Geometry in Computer Vision , R. Hartley and A. Zisserman, Cambridge Press, 2000.
* Tutorial on 3D Modeling from Images , a nice online tutorial which deals with many important practical issues in 3D reconstruction (with some nice demos too).
* Generalized Principal Component Analysis, R. Vidal, Y. Ma, and S. Sastry, (a working manuscript: chapters from the draft will be made available by the instructor.)
Other References (reserved in Grainger, the Engineering Library):
* Geometry of multiple images, O. Faugeras and Q.-T. Luong, MIT press, 2001.
* Principal Component Analysis, I.T. Jolliffe, 2nd Edition, Springer 2002.
*A Mathematical Introduction to Robotic Manipulation, R. Murray, Z.-X. Li and S. Sastry, CRC Press Inc. 1994.
*Three-Dimensional Computer Vision: A Geometric Viewpoint, O. Faugeras, MIT Press, 1993.
*Robot Vision, B. Horn, MIT Press, 1986.
*Theory of Reconstruction from Image Motion, S. Maybank, Springer-Verlag, 1993.
*Motion and Structure From Image Sequences, J. Weng, N. Ahuja and T. Huang, Springer-Verlag, 1993.
*An Introduction to Differential Manifolds and Riemannian Geometry, W. Boothby, Academic Press, 1986.
Grading Policy: Homework (60%) and Final Project (40%).
*Homework: You are allowed to discuss on the homework in small groups, but you must write the solution independently to hand in. No late homework will be accepted (unless an extension is granted by the instructor to the whole class).
*Final Project: The final project can be done in a group of 2 or 3 students - depending on the final size of the class. The project can be theoretical, experimental or a mix of both. It consists of a midterm proposal, a final presentation (in class) and a report. With the instructor's approval, the final project can be related to the student's own graduate research.

Homeworks & Programming Exercises

Handouts & Supplementary Notes

Course Projects

Final project presentation: 2:30 - 4:30pm, EL 245, Thursday, December 9th.
Please make sure your computer is on before the previous presentation is over so that transition is smooth.

Course Projects From Previous Years

Online Resource Links

 Since computer vision is a very active research area, the only way that one can learn about the state of art techniques is to keep tracking what is going on in the world. Once you learn to use the vast resource available outside the classroom, the course itself as well as your own research will most likely become much easier. They are also good places to look for potential final projects or research ideas.
* World computer vision homepage (where, with a little patience, you may find pretty much everything you need to know about vision).
* CVonline collection of computer vision materials (a good tutorial type of online resource for computer vision).
* Common image processing operators (some useful low level image processing routines).
* Basic Image Processing Demos (some old image processing demos that I did a long time ago when I was TAing at Berkeley - the MATLAB codes are already obsolete).
*Optical Illusions (many optical illusions have a lot to do with structure from motion).
* UIUC robotics and computer vision group.
* UC Berkeley computer vision group.
* Stanford computer vision group.
* MIT AI Lab computer vision groups.
* UCLA computer vision group.
* My publication (which may be related to some of the subjects covered in class).

Yi Ma
Last modified: Tue Oct 19 21:27:38 CDT 2004