CAP6412: Advanced Computer Vision - Spring 2007

Meeting Time and Place

Mon. and Wed. 4:30PM - 5:45PM ENG3-0118

Office Hours

TBA

Recommended Texts

Pattern Recognition and Machine Learning - By Christopher Bishop - Link to Book Website - An excellent and thorough introduction to machine learning

Resources

Harris Lab - MATLAB Access - Info

Grading Criteria

Reports - 30% - You will be required to turn in a 1-page report about each paper discussed.
Presentation - 30% - Each course member will be required to present at least one paper
Implementations(2) - 40% - Two implementation projects will be assigned over the course of the semester.

A link to the tentative list of papers can be found here

Current Announcements

01/17/07 - The schedule for papers will be passed out on Monday, Jan 22

Assignment 2

For this project, you have two options. This project will be due April 16th. Please let me know which option you will choose by April 1st.

OPTION 1:

Write a problem set with three problems. Each of the problems should be mathematical in nature and should reinforce an important concept from one of the papers we have written. You should *not* be able to solve any of the problems by simply copying equations from the papers. Examples of acceptable problems would be a derivation that is skipped in the paper or rexpressing a derivation in a different way. You should also turn in solutions to each of the problems.

OPTION 2:

Implement the Spatial-Pyramid Match Kernel. Test it on several object categories from CalTech-101. Note that this project will be much easier to implement if you use code available at: http://people.csail.mit.edu/fergus/iccv2005/bagwords.html I can help with getting libSVM or SVMlight up and running for the classification.

Assignment 1

Implement:

Either the Harris-Affine or Hessian-Affine detector
One other detector, such as the SIFT detector, Kadir and Brady, or the MSER detector. You can find helpful descriptions and routines for the MSER detector at http://vision.ucla.edu/~vedaldi/code/mser/mser.html
No, you cannot turn in the MEX code available there.

All in all, you will implement two detectors. Write a brief report comparing the two and describing what tweaking or magic numbers you had to put in to make it work.

Due Date: Wed, Feb 28, 2007

Course Calendar:

Mon. Jan 8 - Course Administration - Slides
Wed. Jan 10th - Chapter 8: Graphical Models [PDF] Presenter: Dr. Tappen PDF Slides
Mon. Jan 15th - No Class (UCF Holiday - MLK Jr. Day)
Wed. Jan 17th - Ce Liu, Antonio Torralba, William Freeman, Fredo Durand, and Edward Adelson, Motion Magnification, SIGGRAPH 2005 [PDF] Presenter: Jaakko Konttinen
Mon. Jan. 22 - Computing Visual Correspondence with Occlusions using Graph Cuts Vladimir Kolmogorov and Ramin Zabih. In IEEE International Conference on Computer Vision (ICCV), July 2001. - [PDF] Presenter: Min Hu
Wed. Jan 24 - A. Levin and Y. Weiss. Learning to Combine Bottom-Up and Top-Down Segmentation. Proc. of the European Conference on Computer Vision (ECCV), Graz, Austria, May 2006 - [PDF] Presenter: Vladimir Reilly
Monday Jan 29 - R. Fergus, B. Singh, A. Hertzmann, S. Roweis, and W. T. Freeman Removing camera shake from a single image, SIGGRAPH 2006 [PDF] Presenter: Michael Whiteley
Wed. Jan 31 - Fields of Experts: A Framework for Learning Image Priors Stefan Roth and Michael J. Black IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 860-867, San Diego, California, June 2005. - [PDF] Presenter: Remo Pillat
Mon. Feb 05 - A comparison of affine region detectors Krystian Mikolajczyk, Tinne Tuytelaars, Cordelia Schmid, Andrew Zisserman, J. Matas, F. Schaffalitzky, T. Kadir, L. Van Gool International Journal of Computer Vision, Volume 65, Number 1/2 - 2005 - [PDF] Presenter: Justin Key
Wed. Feb 07 - J Portilla, V Strela, M Wainwright, E P Simoncelli. Image Denoising using Scale Mixtures of Gaussians in the Wavelet Domain. IEEE Transactions on Image Processing. vol 12, no. 11, pp. 1338-1351, November 2003.- [PDF] Presenter: Michael Snead
Mon. Feb 12 - Weakly Supervised Scale-Invariant Learning of Models for Visual Recognition - Fergus, R. , Perona, P. and Zisserman, A. - [PDF] Presenter: Jan Prokaj
Wed. Feb 14 - A performance evaluation of local descriptors -Krystian Mikolajczyk, Cordelia Schmid IEEE Transactions on Pattern Analysis & Machine Intelligence, Volume 27, Number 10 - 2005 - [PDF] Presenter: Philip Berkowitz
Mon. Feb 19 -Putting Objects in Perspective Derek Hoiem, Alexei A. Efros, Martial Hebert. In CVPR 2006. - [PDF] Presenter:Pavel Babenko Note Change in Schedule
Wed. Feb. 21 -
K. Grauman and T. Darrell. The Pyramid Match Kernel: Discriminative Classification with Sets of Image Features. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), Beijing, China, October 2005. [PDF]
and
Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories S. Lazebnik, C. Schmid, and J. Ponce. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, New York, June 2006, vol. II, pp. 2169-2178. - [PDF]
Presenter: Jared Johnson
Mon. Feb 26 - D. Blei, A. Ng, and M. Jordan. Latent Dirichlet allocation. Journal of Machine Learning Research, 3:993–1022, January 2003. - [PDF] Presenter: Kegan Samuel
Wed. Feb 28 - J. Sivic, B. C. Russell, A. A. Efros, A. Zisserman, W. T. Freeman, Discovering objects and their location in images, International Conference on Computer Vision (ICCV), Beijing, China, October, 2005.- [PDF]
and
J. Sivic, B. C. Russell, A. A. Efros, A. Zisserman, W. T. Freeman, Discovering object categories in image collections, MIT AI Lab Memo AIM-2005-005, February, 2005 - [PDF] Presenter: Yusuf Ayatar
Mon. March 5 - Continuation of previous lecture
Wed. March 7 - The earth mover's distance as a metric for image retrieval. International Journal of Computer Vision, 40(2):99-121, November 2000 Y. Rubner, C. Tomasi, and L. J. Guibas. [PDF] Presenter: Sohail Sattar
Mon. March 19 - Context-based vision system for place and object recognition A. Torralba, K. P. Murphy, W. T. Freeman and M. A. Rubin IEEE Intl. Conference on Computer Vision (ICCV), Nice, France, October 2003. - [PDF] Presenter: Ryan Faircloth
Wed. March 21 - Contextual Models for Object Detection using Boosted Random Fields A. Torralba, K. P. Murphy and W. T. Freeman Adv. in Neural Information Processing Systems 17 (NIPS), pp. 1401-1408, 2005. - [PDF] Presenter: Mikel Rodriguez
Mon. March 26 - Shape Matching and Object Recognition using Low Distortion Correspondence Alexander C. Berg, Tamara L. Berg, Jitendra Malik - IEEE Computer Vision and Pattern Recognition (CVPR) 2005 - [PDF] Presenter: David Lyle
Mon. April 2 - Sharing visual features for multiclass and multiview object detection A. Torralba, K. P. Murphy and W. T. Freeman In press, IEEE Transactions On Pattern Analysis and Machine Intelligence. - [PDF] Presenter: Craig Dean
Wed. April 4 >Jian Sun, Yin Li, Sing-Bing Kang and Heung-Yeung Shum. Symmetric Stereo Matching for Occlusion Handling. CVPR 2005, Vol. 2, pp. 399-406. [PDF] Presenter: Omer Bilal
Mon. April 9 - J.C. Niebles, H. Wang, L. Fei-Fei. Unsupervised learning of human action categories using spatial-temporal words. [PDF] -Presenter: Paul Sovanner
Wed. April 11 Presenter: Vladimir Reilly (His CVPR 2007 paper) [PDF]
Mon. April 16 - Space-Time Behavior Based Correlation [Website] Presenter: Jin-Gen Liu
Wed. April 18 - Multiscale CRFs for Image Labeling - [PDF] Presenter: Pavel Babenko
Mon. April 23 - Learning Gaussian Conditional Random Fields - [PDF] No Report is Necessary Today