Readings, lecture notes, and computer examples covered so far:

• Monday, Feb. 11.
  Paper: Emergence of simple-cell receptive field properties by learning a sparse code for natural images, Olshausen BA, Field DJ (1996) Nature, 381: 607-609. (pdf file)
  Presenter: Bill Freeman
  • (lecture notes in html),
  Computer example presenter: Thomas Serre
  • (ppt),
  • (computer code used for demonstrations),
  Neat videos of Hubel and Wiesel Via Ben Pearre
  • (link),
  
  
  
• Wednesday, Feb. 13.
  Paper: Learning the parts of objects by non-negative matrix factorization, D. D. Lee and H. S. Seung, Nature 401, 788-791 (1999). (pdf)
  
  • Also this commentary: B. W. Mel. Computational neuroscience: Think positive to find parts, News and Views, Nature, 401, 759-760 (1999). (pdf)
  • Optional: this paper describes some details not given in the Nature paper: Algorithms for non-negative matrix factorization, D. D. Lee and H. S. Seung, NIPS 13, pp. 556-562 (2001). (pdf)
  • A follow-on paper by other authors, explicitly enforces locality constraint in NMF framework: "Learning [a] spatially localized, parts-based representation", by Li, Hou, Zhang, and Cheng, IEEE Computer Vision and Pattern Recognition (CVPR) 2001. (pdf)
  Presenter: Marshall Tappen
  
  • (ppt),
  Some comments: Bill Freeman
  
  • (ppt),
  Computer example presenter: Bryan Russell
  • (ppt),
  • (tar file of examples),
  
  
  
• Tuesday Feb. 19.
  Paper: Nonlinear dimensionality reduction by locally linear embedding. Sam Roweis & Lawrence Saul. Science v.290 no.5500, Dec.22, 2000. pp.2323--2326. (website) (paper)
  
  Related paper on manifold learning:
  • Surface learning with applications to lipreading, C. Bregler and S. Omohundro, NIPS (Advances in Neural Information Processing Systems 6, Morgan Kaufmann Publishers), 1994 (pdf)
  Presenter: Kinh Tieu
  Slides on Bregler & Omohundro: Bill Freeman
  
  • (ppt),
  Computer example presenter: Kinh Tieu
  • (pdf of both of Kinh's presentations),
  • Data sets he used (CBCL data sets of faces, cars, pedestrians)
  
  
  
• Weds., Feb. 20.
  Paper: A global geometric framework for nonlinear dimensionality reduction J. B. Tenenbaum, V. De Silva, J. C. Langford Science 290 (5500): 22 December 2000. (website) (paper)
  Presenter: Josh Tenenbaum
  Computer example presenter: Matt Grimes
  • (ppt),
    
  • "Here's the matlab code I used to generate my movies for the Isomap demo".
    
  
  
  
• Monday, Feb. 25. Room 24-121.
  Paper: S. Soatto, G. Doretto, Y.Wu Dynamic Textures. Intl. Conf. on Computer Vision, pages 439-446, July 2001. (link)
  Optional, but recommended, second reading: Fitzgibbon, A. W. Stochastic rigidity: Image registration for nowhere-static scenes Proc. International Conference on Computer Vision, July, 2001. (pdf)
  Presenter: Erik Sudderth
  Computer example presenter: Erik Sudderth
  For his presentation: All of the data is posted in the following directory: (link) .
  There are 3 archives for the 3 video examples, plus the powerpoint file and an archive containing Matlab code.
  
  
• Wednesday, Feb. 27. Room 24-121
  Paper: Separating style and content with bilinear models. J. B. Tenenbaum, W. T. Freeman (2000) Neural Computation 12 (6), 1247-1283. (pdf)
  Presenter: Bill Freeman
  
  • (ppt),
    
  Computer example presenter: Barun Singh
  • (ppt),
    
  
  
  
• Monday, March 4.
  Paper: Deriving intrinsic images from image sequences Weiss Y. International Conference on Computer Vision (ICCV) 2001, Vancouver, BC, Canada. (pdf)
  Presenter: Leonid Taycher
  • (ppt),
  Computer example presenter: Peter Sand
  • (ppt),
    
  
  
  
• Wednesday, March 6.
  Paper: Comparison of statistical image models for image processing applications, Eero Simoncelli.
  Readings: Two short conference papers, about statistical image models for texture synthesis and image denoising: (paper 1) (paper 2)
  Presenter: Eero Simoncelli
  Eero's presentation slides:
  (pdf1) and (pdf2)
  
  
• Monday, March 11.
  Paper: R. Szeliski. Bayesian modeling of uncertainty in low-level vision. International Journal of Computer Vision, 5(3):271-301, December 1990
  Comment: Classic paper about Bayesian methods.
  Presenter: Michael Ross
  His slides (pdf)
  Bill's slides: (ppt),
  Computer example presenter: Bhiksha Ramakrishnan His slides (ppt)
  
  
• Wednesday, March 13.
  Paper: Interpreting images by propagating Bayesian beliefs. Weiss Y. in: M.C. Mozer, M.I. Jordan and T. Petsche, editors, Advances in Neural Information Processing Systems 9 908-915 (1997). (pdf)
  Comment: Here we derive and cover belief propagation.
  Presenter: David Rosenberg (ppt)
  Bill Freeman's slides
  Computer example presenter: David Rosenberg
  
  
• Monday, March 18.
  Paper: Learning low-level vision, W. T. Freeman and E. C. Pasztor, Intl. Conf. on Computer Vision, Corfu, Greece, 1999. (pdf)
  
  Optional paper: longer, with more details about the belief propagation.
  • Learning low-level vision, W. T. Freeman and E. C. Pasztor and O. T. Carmichael, Intl. J. Computer Vision, vol. 40, no 1, pp. 25 -- 47, 2000. (pdf)
  Presenter: Bill Freeman (ppt)
  Computer example presenter: Michael Ross (ppt)
  
  
• Wednesday, March 20.
  Paper: Fitting parameterized three-dimensional models to images David G. Lowe, IEEE Transactions on Pattern Analysis and Machine Intelligence, 13, 5 (May 1991), pp. 441-450. [HTML] [PDF]
  Comment: Classic object-based recognition.
  Presenter: Ray Jones (pdf)
  Bill's slides about the course: (ppt)
  Computer example presenter: Ray Jones (pdf)
  Ray's code for his toy example: (link)
  
  
  Recommended second reading:
  Parallel integration of vision modules, T. Poggio, E. B. Gamble, J. J. Little, Science, vol. 242, pp. 337--484, October 21, 1988. (pdf)
  Presenter: Brian Whitman (ppt)
  
  
• Monday, April 1.
  Paper:
  The short version:
  Probabilistic Object Recognition and Localization. Bernt Schiele and Alex Pentland. In ICCV'99 International Conference on Computer Vision. (pdf)
  Ok to read just the above, but for more info, read the long version:
  Recognition without Correspondence using Multidimensional Receptive Field Histograms. Bernt Schiele and James L. Crowley. In International Journal of Computer Vision 36 (1), p 31-50, January 2000. (pdf)
  Presenter: Matt Grimes
  His slides: (ppt)
  Computer example presenter: Marshall Tappen (ppt)
  
  
• Wednesday April 3.
  Paper: Bernd Heisele, Thomas Serre, Massimiliano Pontil, Thomas Vetter, and Tomaso Poggio, Categorization by learning and combining object parts. (pdf)
  Presenter: Manish Jethwa
  Computer example presenter: Manish Jethwa (ppt)
  
  
• Monday, April 8.
  Paper: Robust real-time object detection, Paul Viola and Michael Jones, 2nd intl workshop on statistical and computational theories of vision, June, 2001. (compressed postscript)
  Presenter: Paul Viola
  Computer example presenter: Laptop demo by Paul Viola.
  
  
• Wednesday, April 10.
  Talk on "How to write a conference paper"
  Papers:
  • How to Get Your SIGGRAPH Paper Rejected, Jim Kajiya, SIGGRAPH 1993 Papers Chair, (link)
  • Ted Adelson's Informal guidelines for writing a paper, 1991. (link)
  • Notes on technical writing, Don Knuth, 1989. (pdf)
  • What's wrong with these equations, David Mermin, Physics Today, Oct., 1989. (pdf)
  • Ten Simple Rules for Mathematical Writing, Dimitri P. Bertsekas (link)
  
  Presenter: Bill Freeman (.ppt)
  
  
• Wednesday April 17.
  Papers (read one or both of these):
  • Interpreting face images using active appearance models, Edwards, Taylor and Cootes, Conference on Face and Gesture Recognition, 1998. (.pdf)
  • Constrained Active Appearance Models, T.F.Cootes and C.J.Taylor, Proc. ICCV 2001. Vol.I. pp748-754 (gzipped postscript)
  Presenter: Byran Russell
  His presentation: (.ppt)
  Computer example presenter: Antonio Torralba
  There is a free downloadable implementation of the AAM by Mikkel Stegman. (link)
  Antonio's presentation: (ppt)
  His matlab code: (zip)
  
  
  
• Monday April 22.
  Presenter: Baback Moghaddam
  Comment: Baback will focus on Bayesian face-recognition and its pre-requisite: PCA-based density estimation. He'll also talk a bit about using SVM's for gender recognition, too, as an application idea and as an illustration of other types of facial processing (classification as opposed to recognition).
  Paper: Probabilistic Visual Learning for Object Detection, B. Moghaddam and A. Pentland, 5th Intl. Conf. on Computer Vision, 1995. (pdf)
  Optional paper: Learning Gender with Support Faces, B. Moghaddam and M-H. Yang, IEEE Transactions on Pattern Analysis & Machine Intelligence, Vol. 64, No. 5, May 2002. (pdf)
  Computer example presenter: Brian Whitman
  
  
• Wednesday, April 24.
  Paper: Object recognition as machine translation--I: Learning a lexicon for a fixed image vocabulary, Pinar Duygulu, Kobus Barnard, Nando deFreitas, and David Forsyth, 2002. (pdf) .
  There's also an optional companion paper, (pdf) .
  Comment: They look at a large database of images labelled with text, and try to learn how to "translate" from text to parts of the images.
  Presenter: Charlie Kemp
  Computer example presenter: Charlie Kemp (ppt)
  
  
• Monday, April 29.
  Paper: Transformed hidden Markov models: Estimating mixture models of images and inferring spatial transformations in video sequences. N. Jojic, N. Petrovic, B. J. Frey and T. S. Huang. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2000, IEEE Computer Society Press, Los Alamatos, CA. (link)
  Presenter: Barun Singh
  
  
• Wednesday May 1.
  Paper: Markus Weber, Max Welling & Pietro Perona (2000) Unsupervised Learning of Models for Recognition Proc. 6th Europ. Conf. Comp. Vis., ECCV2000, Dublin (gzipped postscript)
  Presenter: Gregory Shakhnarovich
  Computer example presenter: Gregory Shakhnarovich
  
  
• Monday May 6.
  Paper: Blake, A, Isard, M, and Reynard, D., "Learning to track the visual motion of contours", Artificial Intelligence Journal, 1995. (pdf)
  Presenter: Andrew Blake
  
  
• Wednesday May 8.
  Paper: Contour tracking by stochastic propagation of conditional density Michael Isard and Andrew Blake Proc. European Conf. on Computer Vision, vol. 1, pp. 343--356, Cambridge UK, (1996). (pdf).
  Presenter: Andrew Blake
  Computer example presenter: Yuan Qi
  
  
• Monday May 13.
  Papers (read one and skim the other; your choice): Sidenbladh, H. and Black, M. J., Learning image statistics for Bayesian tracking, Int. Conf. on Computer Vision, ICCV-2001, Vancouver, BC. (postscript, 2.8MB)(pdf, 0.38MB). (link)
  Implicit probabilistic models of human motion for synthesis and tracking, Sidenbladh, H., Black, M. J., and Sigal, L., to appear: European Conf. on Computer Vision, ECCV2002. (pdf)
  Presenter: Yajun Fang
  Computer example presenter: Yajun Fang
  
  
• Wednesday May 15.
  Paper: Jepson, A.D., Fleet, D.J. and El-Maraghi, T. (2001) Robust, on-line appearance models for vision tracking. IEEE Conference on Computer Vision and Pattern Recognition, Kauai, Vol. I, pp. 415--422 (pdf)
  Presenter: Peter Sand
  Computer examples by: David Fleet