Readings

Belief Propagation

• Michael I Jordan and Yair Weiss, Probabilistic inference in graphical models, book chapter
• William T. Freeman, Egon C. Pasztor, Learning low-level vision, IJCV 2000
• M. F. Tappen and W. T. Freeman, Comparison of graph cuts with belief propagation for stereo, using identical MRF parameters, ICCV 2003
• (optional) Pedro Felzenszwalb and Dan Huttenlocher, Efficient Belief Propagation for Early Vision, to appear in CVPR, 2004 .
• (optional) Nonparametric belief propagation (for continuous random variables): http://ssg.mit.edu/nbp/, also related work by Michael Isard.
• (optional) Sun, H. Y. Shum, and N. N. Zheng. Stereo matching using belief propagation. PAMI 2003
• (optional) James M. Coughlan and Sabino J. Ferreira, Finding Deformable Shapes using Loopy Belief Propagation, ECCV 2002.

Variational Inference and Expectation Maximization

• Brendan J. Frey and Nebojsa Jojic, Advances in algorithms for inference and learning in complex probability models, tutorial
• N. Jojic, B.Frey & A.Kannan , Epitomic analysis of appearance and shape, ICCV 2003
• N. Jojic, and B.Frey, Learning Flexible Sprites in Video Layers, CVPR 2001
• (optional) P. H. S. Torr, R. Szeliski, and P. Anandan. An integrated Bayesian approach to layer extraction from image sequences. IEEE PAMI, March 2001
• (optional) N. Jojic, and B.Frey, Transformation-Invariant Clustering Using the EM Algorithm, IEEE PAMI, Jan 2003
• (optional) Tutorials: http://www.cs.brown.edu/research/ai/dynamics/tutorial/Documents/ExpectationMaximization.html, and http://www.cc.gatech.edu/~dellaert/em-paper.pdf.
• (optional) Y. Weiss and E.H. Adelson, Perceptually organized EM: A framework for motion segmentation that combines information about form and motion, MIT Media Lab Perceptual Computing Section TR #315, 1994
• (optional) L. Torresani, A. Hertzmann, Automatic Non-Rigid 3D Modeling from Video, errata, Proc. ECCV 2004

Graph Cuts

• Olga Veksler, Efficient Graph-Based Energy Minimization Methods in Computer Vision, Ph.D. thesis (read chapters 1, 2)
• Stan Birchfield and Carlo Tomasi, Multiway Cut for Stereo and Motion with Slanted Surfaces, ICCV 1999
• Dan Snow, Paul Viola and Ramin Zabih, Exact Voxel Occupancy with Graph Cuts, CVPR 2000
• graph cuts home page, including pointers to software, lots of papers
• min cut/max flow link
• (optional) V. Kwatra , A. Schödl , I. Essa , G. Turk and A. Bobick, Graphcut textures: image and video synthesis using graph cuts, SIGGRAPH 2003
• (optional) The Boykov-Kolmogorov PAMI paper is good for someone who wants to implement graph cuts.  Ramin suggests reading sections 1 and 2. If they want to implement something, I (Ramin) would definitely do an augmenting paths algorithm, either Ford-Fulkerson or Edmonds-Karp. You can find good descriptions of these from, for example, Google (http://www.cis.ksu.edu/~howell/775f01/slides/slides18-01.pdf is one example)
• (optional) The Boykov-Veksler-Zabih PAMI paper from 2001. This is the definitive paper, but not easy reading. However, you can skip all the proofs without harm.
• (optional)  Kolmogorov & Zabih ECCV 2002 paper.  Very cool use of graph cuts for multiview stereo, handling visibility.
• (optional) Kolmogorov-Zabih PAMI 04, Theoretical but important paper, characterizes energy functions that can be minimized by graph cuts.

Level Sets

• G. Sapiro, Geometric partial Differential Equations and Image Analysis, Cambridge University Press, Chapter 2, 3, and 4
• R. Kimmel, Numerical Geomtry of Images, Chapter 5
• (optional) M. Bertalmío, G. Sapiro, V. Caselles and C. Ballester. Image Inpainting, Proceedings of SIGGRAPH 2000
• (optional) V. Caselles, R. Kimmel, and G. Sapiro, Geodesic active contours, International Journal of Computer Vision, pp. 61-79, 1997
• (optional) B. Tang, G. Sapiro, and V. Caselles, Diffusion of general data on non-flat manifolds via harmonic maps theory: The direction diffusion case, Int. Journal Computer Vision, pp. 149-161, February 2000
• (optional) F. Memoli and G. Sapiro, Fast computation of weighted distance functions and geodesics on implicit hyper-surfaces, Journal of Computational Physics, pp. 730-764, November 2001
• (optional) M. Bertalmio, G. Sapiro, and G. Randall, Morphing active contours, IEEE Trans. Pattern Analysis Machine Intelligence, pp. 733-738, 2000
• (optional) M. Bertalmio, G. Sapiro, and G. Randall, Region tracking on level-sets methods, IEEE Trans. Medical Imaging, pp. 448-451, 1999
• (optional) G. Sapiro, R. Kimmel, D. Shaked, B. Kimia, and A. M. Bruckstein. Implementing continuous-scale morphology via curve evolution. Pattern Recognition, 26(9):1363-1372, 1993
• (optional) L. Cohen and R. Kimmel. Global minimum for active contours models: A minimal path approach. International Journal of Computer Vision, 24(1):57-78, 1997
• (optional) R. Kimmel and J. A. Sethian. Computing Geodesic Paths on Manifolds, Proceedings of National Academy of Sciences, 95(15):8431-8435, July, 1998
• (optional) Olivier Faugeras and Renaud Keriven, Variational Principles, Surface Evolution, PDE's, level set methods and the Stereo Problem, Special Issue on Partial Differential Equations and Geometry-Driven Diffusion in Image Processing and Analysis of the IEEE Transactions on Image Processing, Vol. 7, No. 3, pages 336-344, March 1998
• (optional) Gerardo Hermosillo, Olivier Faugeras and Jos Gomes, Cortex Unfolding Using Level Set Methods, Inria Research Report 3663, June 1999.

Non-Linear Least Squares and Sparse Matrix Techniques

• Press et al, Numerical Recipes, Chapter 15 (Modeling of Data)
• (optional) Nocedal and Wright, Numerical Optimization, Chapter 10 (Nonlinear Least-Squares Problems, pp. 250-273)
  Press et al is a gentler introduction, and also talks about robust regression. Nocedal and Wright contains more details on what actually works. However, Nocedal and Wright is not available online, you'd have to check it out from the library.
• Shewchuk, J. R. An Introduction to the Conjugate Gradient Method Without the Agonizing Pain.
• R. Szeliski and S. B. Kang. Recovering 3D shape and motion from image streams using nonlinear least squares. Journal of Visual Communication and Image Representation, 5(1):10-28, March 1994. (Sparse non-linear LS applied to structure from motion.) Also available as CRL-93-3
  *or* R.Szeliski. Fast surface interpolation using hierarchical basis functions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(6):513-528, June 1990. (Hierarchical basis pre-conditioning of conjugate gradient.)
• (optional)Bathe and Wilson, Numerical Methods in Finite Element Analysis, Prentice-Hall, 1976. Description of skyline storage and sparse LDU factorization, page 695-717 (sections 8.1-8.2) and page 979-987 (section 12.2).
• (optional)Golub and VanLoan, Matrix Computations. Chapters 4, 5, and 10.
• (optional)Nocedal and Wright, Numerical Optimization. Chapters 4 and 5.
• (optional)B. Triggs, P. McLauchlan, R. Hartley & A. Fitzgibbon, Bundle Adjustment -- A Modern Synthesis (Revised version to appear in final proceedings of Vision Algorithms'99).
  

Discriminitive Methods

• Robert E. Schapire, Yoram Singer, Improved Boosting Algorithms Using Confidence-rated Predictions, Computational Learing Theory
• Paul Viola, Michael Jones, Robust Real-time Object Detection, International Journal of Computer Vision 2001
• (optional) Viola, P.; Jones, M.J.; Snow, D., Detecting Pedestrians Using Patterns of Motion and Appearance, IEEE International Conference on Computer Vision (ICCV), pp. 734-741, October 2003
• (optional) Jianxin Wu James M. Rehg Matthew D. Mullin, Learning a Rare Event Detection Cascade by Direct Feature Selection,
• (optional) Antonio Torralba Kevin P. Murphy William T. Freeman, Sharing features: efficient boosting procedures for multiclass object detection, CVPR 2004
• (optional) Kevin Murphy, Antonio Torralba, William T. Freeman, Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes, NIPS 2003
• (optional) Michael Fink, Pietro Perona, Mutual Boosting for Contextual Inference, NIPS 2003

Dimensionality Reduction

• Sam Roweis, Lawrence Saul, Nonlinear Dimensionality Reduction by Locally Linear Embedding, Science v.290 no.5500, Dec.22, 2000. pp.2323--2326
• Joshua B. Tenenbaum, Vin de Silva, John C. Langford, A Global Geometric Framework for Nonlinear Dimensionality Reduction, Science
• Kilian Q. Weinberger and Lawrence K. Saul, Unsupervised Learning of Image Manifolds by Semidefinite Programming, CVPR 2004
• (optional) Yoshua Bengio, Jean-Franc¸ois Paiement and Pascal Vincent, Out-of-Sample Extensions for LLE, Isomap, MDS, Eigenmaps, and Spectral Clustering, Technical Report 1238
• (optional) Jihun Ham, Daniel D. Lee, Sebastian Mika, Bernhard Scholkopf, A kernel view of the dimensionality reduction of manifolds, Technical Report TR-110

Distance Transform & Matching

• Pedro F. Felzenszwalb, and Daniel P. Huttenlocher, Distance Transforms of Sampled Functions
• (optional) Pedro Felzenszwalb and Dan Huttenlocher, Efficient Belief Propagation for Early Vision, to appear in CVPR, 2004.

MCMC

• Introduction to Monte Carlo methods, A review paper in the proceedings of an Erice summer school, ed. M.Jordan.
• Frank Dellaert, A Sample of Monte Carlo Methods in Robotics and Vision.
• F. Dellaert , S. Seitz , C. Thorpe , and S. Thrun, EM, MCMC, and Chain Flipping for Structure from Motion with  Unknown Correspondence. Machine Learning, special issue on Markov chain Monte Carlo methods., 50, pp. 45-71, 2003.
• Z. Khan, T. Balch, and F. Dellaert, An MCMC-based Particle Filter for Tracking Multiple Interacting Targets, European Conference on Computer Vision (ECCV 04), 2004.