Term: Spring 2017-2018
CSC 249/449 Machine Vision (Spring 2017)
Fundamentals of computer vision, including image formation, elements of human vision, low-level image processing, and pattern recognition techniques. Advanced topics include modern visual features, graphical models, model-based and data-driven approaches, and contextual inference, as well as illustrative examples of successes and challenges in applications. CSC 449, a graduate-level course, requires additional readings and assignments (including a course project). Based on past classes, CSC 249, an undergraduate-level course, has exactly the same requirements with a 5% bonus for completing the same requirements.
Please check pre-requisites on the official course listing page. Some knowledge of linear algebra and basic statistics is helpful but we will review these subjects in the class (please discuss with me if you are not sure). For assignments and projects, the most common programming languages are Matlab, C or C++. Note that Matlab provides an image processing toolbox and is available on most CS department undergraduate and research networks.
Course schedule (tentative - subject to changes)
|week||dates||chapter||key topics||CSC 249||CSC 449||notes|
|1||1/19||1, 2||overview; image formation|
|2||1/24, 1/26||3, 4||color vision; low-level image processing|
|3||1/31, 2/2||4, notes (Forsyth)||texture; Intro to (review of) probability|
|4||2/7, 2/9||Duda & Hart||pattern recognition concepts (incl. classifiers)|
|5||2/14, 2/16||5, 6||modern visual features|
|6||2/21, 2/23||9, 10||mid-level vision (incl. segmentation, Hough Transform)|
|notes & special lecture||recognizing faces and people|
|8||3/7, 3/9||review/midterm exam|
|9||3/14, 3/16||Spring Break, no class|
|10||3/21, 3/23||midterm solutions, 12||image registration||selecting course project topics||selecting course project topics||2-page proposal|
|11||3/28, 3/30||guest lectures||deep learning in computer vision; vision and language|
|12||4/4, 4/6||11, 7, 8||tracking, stereo & 3D|
|13||411, 4/13||18, notes||actions and events|
|14||4/18, 4/20||notes||data-driven computer vision|
|15||4/25, 4/27||21, notes & special lecture||content-based image annotation & retrieval (or alternatively medical image analysis)||or attribute learning in vision|
|16||5/2||notes & special lecture||contextual inference in computer vision|
|16||5/4||final presentations & reports||impress us!||5/4|
|17||5/9, 5/11||final presentations & reports||impress us!||5/9 & 5/11||5/9 & 5/11|
Instructor and grading
Instructor: Prof. Jiebo Luo, CSB Rm 611, x65784
TA: Yang Feng (CSB 603)
Lectures: Tuesdays and Thursdays, 12:30-13:45, CSB 601
Office hours: after classes (15:00-16:00) or by appointment (use email).
Grading (total 100%)
- midterm 30%
- homework assignments 35% (including a small project 10%)
- final project & presentation (for CSC 449) 30% (presentation counts 10%)
- class participation and effort 5%
- an existing algorithm applied to new data or new problems
- a new algorithm (or a modified version of an existing algorithm) applied to the same data
- new findings from a comparative study of using different algorithms for the same problem
Textbooks and other resources
Computer Vision: A Modern Approach, 2/E
Publisher: Prentice Hall, 2012
Pattern Classification, 2/E
Richard O. Duda, Peter E. Hart and David G. Stork
Publisher: John Wiley & Sons, 2001
Recommended papers (use IEEE Xplore on the university network for free access to these and other papers):
Longuet-Higgins Prize recipients
(CVPR annual award for 'fundamental contributions in Computer Vision that Have Withstood the Test of Time')
- Ted Adelson and John Wang, "Layered representation for motion analysis", CVPR 1993.
- David Mumford and Jayant Shah, "Boundary detection by minimizing functionals", CVPR 1985.
- H. Rowley, S. Baluja and Takeo Kanade, "Neural Network-Based Face Detection", CVPR 1996.
- Cordelia Schmid and Roger Mohr, "Combining greyvalue invariants with local constraints for object recognition", CVPR 1996.
- J. Shi and J. Malik, "Normalized Cuts and Image Segmentation", CVPR 1997.
- E. Osuna, R. Freund, and F. Girosi, "Training Support Vector Machines: An Application to Face Detection", CVPR 1997.
- H. Schneiderman and Takeo Kanade, "Probabilistic modeling of local appearance and spatial relationships for object recognition", CVPR 1998.
- C. Bregler and J. Malik, "Tracking people with twists and exponential maps", CVPR 1998.
- Jinggang Huang, David Mumford, "Statistics of Natural Images and Models", CVPR 1999.
- Chris Stauffer, W.E.L. Grimson, "Adaptive Background Mixture Models for Real-Time Tracking", CVPR 1999.
- Pedro F. Felzenszwalb and Daniel P. Huttenlocher, "Efficient Matching of Pictorial Structures", CVPR 2000.
- Dorin Comaniciu, Visvanathan Ramesh, and Peter Meer, "Real-Time Tracking of Non-Rigid Objects Using Mean Shift", CVPR 2000.
- Paul A. Viola and Michael J. Jones, "Rapid Object Detection using a Boosted Cascade of Simple Features", CVPR 2001.
- 2012 – break