Computer Vision: A Modern Approach
by Forsyth and Ponce
There will also be a number of supplemental readings
Course Overview
This course will focus on topics in and
applications of level computer vision. While we will cover a number
of areas of in computer vision, we will focus on high-level computer
vision, examining techniques that enable the computer to perform
automatic object segmentation and recognition in both visual and
sketched images. Through a number of programming projects, we will
examine techniques including binary and color image analysis, object
segmentation, model representation, pattern classification, and sketch
processing and recognition.
Links
Calendar
This calendar will be updated as the class progresses. I will try to
post reading assignments at least a day or two in advance of the relevant
lecture.
| Date | Lecture | Reading | Homework |
| Wed, Aug 31 | Introduction | Chapter 1 | PS 1 Out |
| Mon, Sept 5 | Light and Sources | Chapter 4, 5.2 | |
| Wed, Sept 7 | Binary Images and Filtering | N/A | PS 1 Due (midnight) PS 2 out |
| Mon, Sept 12 | Color, Reading a research paper | Chapter 6 | |
| Wed, Sept 14 | Segmentation by clustering, Filters | Chapter 14.1-14.2, 14.4, 7.1-7.3,7.5 | |
| Mon, Sept 19 | Research paper #1, Texture | Contour
and Texture Analysis for Image Segmentation Chapter 9.1-9.2 | Paper review |
| Wed, Sept 21 | Intro to Recognition by Classification | Chapter 22.1 | PS 2 Due on Friday |
| Mon, Sept 26 | More Classification and PCA | Chapter 22.2-22.4 | |
| Wed, Sept 28 | More PCA | PS 3 out | |
| Mon, Oct 3 | Paper presentation #2 | Eigenfaces for Recognition by Turk and Pentland [TP91] | Paper review #2 due |
| Wed, Oct 5 | Line/Shape Fitting via the Hough Transform | Chapter 15.1-15.3 | |
| Mon, Oct 10 | More Line/Shape Fitting Probabilistic Models | 15.4-15.5 | |
| Wed, Oct 12 | Probabilistic Fitting (cont) | 15.4-15.5, Ch 16 (EM) | PS 3 Due on Friday 10/14 by 5pm |
| Mon, Oct 17 | No Class (Fall break) | ||
| Wed, Oct 19 | EM (cont) Paper presentation #3 | Ch 16 (EM) Contextual Priming for Object Detection by A. Torralba | Paper review #3 due |
| Mon, Oct 24 | NO CLASS | ||
| Wed, Oct 26 | More Complex Model Fitting Challenges of Sketch Understanding | Ch 18 (some material not in book) | Project Proposal due on Friday 10/28 by 11:59pm PS4 Out |
| Mon, Oct 31 | Paper presentation #4 | Combining geometry and domain knowledge to interpret hand-drawn diagrams by Gennari, Kara, Stahovich and Shimada | Paper review due |
| Wed, Nov 2 | Stroke Parsing | [CD04], [S04], [SD04] | |
| Mon, Nov 7 | Camera Parameters and Calibration | Ch 2, 3.1 | |
| Wed, Nov 9 | Stereopsis | Ch 10.1, 11 | PS 4 Due on Friday 11/11 by 11:59pm |
| Mon, Nov 14 | Model-Based Vision | Ch 18 | |
| Wed, Nov 16 | Model-Based Vision (cont) | Ch 18 | PS 4 Extension and redo of part 2 due Friday |
| Mon, Nov 21 | Paper Presentation #5 | "Utilizing Segmented MRI Data in Image-Guided Surgery" by Grimson et al. | Paper review #5 |
| Wed, Nov 16 | Project Meetings |
[MBLS01] Jitendra Malik, Serge Belongie, Thomas Leung and Jianbo Shi. Contour and Texture Analysis for Image Segmentation. International Journal of Computer Vision 43(1). pp. 7-27.
[TP91] Matthew Turk and Alex Pentland. "Eigenfaces for Recognition." Journal of Cognitive Neuroscience, 3 (1). 1991.
[T03] Antonio Torralba. "Contextual Priming for Object Detection". International Journal of Computer Vision, 53 (2): 153-167.
[GKSS05] Leslie Gennari, Levent Burak Kara, Thomas F. Stahovich, and Kenji Shimada. Combining geometry and domain knowledge to interpret hand-drawn diagrams. Computers and Graphics 29 (4) 2005.
[CD05] Sonya Cates and Radall Davis, A New Approach to Early Sketch Processing. AAAI Symposium on Making Pen-Based Interfaces Intelligent and Natural. 2004.
[S05] Thomas F. Stahovich. Segmentation of Pen Strokes Using Pen Speed. AAAI Symposium on Making Pen-Based Interfaces Intelligent and Natural. 2004.
[SD05] Tevfik Metin Sezgin and Radall Davis, Scale-Space Feature Point Detection for Digital Ink. AAAI Symposium on Making Pen-Based Interfaces Intelligent and Natural. 2004.
[GEKLWK97] E. Grimson, G. Ettinger, T. Kapur, M. Leventon,
W. Wells, R. Kikinis, "Utilizing Segmented MRI Data in Image-Guided
Surgery", Int. J. Pattern Recognition & Artificial Intelligence,
special issue on Processing, Analysis, and Understanding of Magnetic
Resonance Images of the Human Brain, 11(8):1367-1397, 1997
If you have any question at all as to what is considered
acceptable collaboration, please ask.
Honor Code
Your work in this class should be in conformance with the Harvey Mudd
honor code. For the problem sets, you may discuss problems with other
students, but you should each write up/program your own solutions.
Please list names of all people with whom you discussed problems in
your writeup. For the project, you will be given the opportunity to
work in teams of two or three students, but it is important that the
work submitted be that only the students in the group, as well as the
work of all of the students in the group. For the paper
reviews, while you may discuss the papers, each student must write his
or her own review.