Instructor: Jason Corso (UBIT: jcorso)

Course Webpage: http://www.cse.buffalo.edu/~jcorso/t/CSE672.

Syllabus: http://www.cse.buffalo.edu/~jcorso/t/CSE672/files/syllabus.pdf.

Downloadable course material can be found on the CSE UNIX network: /home/csefaculty/jcorso/672.

Meeting Times: TR 12:30-1:50

Location: Fronczak Hall 422 (http://goo.gl/maps/0Y4mz)

Email Listserv: cse672-fa12-list@listserv.buffalo.edu
Use this list for any and all course discussion, except private matters.

News

9/4 -- Project idea examples in /home/csefaculty/jcorso/672/examples.pdf on the CSE network.
8/28 -- First day of class.

Calendar

The calendar is given in weeks and will be populated as the semester proceeds based on the course outline and our progress. There are no slides for this course (lectures are given on the board) and you should cross-reference reading materials with the outline below and the bibliography I handed out with the syllabus.

August 28: Introduction. Statistics of Natural Images.
Sept. 4: Statistics of Natural Images wrap-up.; Descriptive models on regular lattices.
Sept. 11: Project Proposal Due in Class on 9/11; Descriptive models on regular lattices.
Sept. 18: Project Proposal Due in Class on 9/20
Sept. 25
Oct. 2: Project Milestone 1 Report Due in Class on 10/4
Oct. 9
Oct. 16: Project Milestone 2 Report Due in Class on 10/18
Oct. 23
Oct. 30: Project Milestone 3 Paper Due in Class on 11/1
Nov. 6: Peer Reviews Due 11/6
Nov. 13: Revised project paper due 11/15
Nov. 20
Nov. 27
Dec. 4

Main Course Material

Course Overview: The course takes an in-depth look at various Bayesian methods in computer and medical vision. Through the language of Bayesian inference, the course will present a coherent view of the approaches to various key problems such as detecting objects in images, segmenting object boundaries, and recognizing activities in video. The course is roughly partitioned into two parts: modeling and inference. In the first half, it will cover both classical models such as weak membrane models and Markov random fields as well as more recent models such as conditional random fields, and topic models. In the second half, it will focus on inference algorithms. Methods include PDE boundary evolution algorithms such as region competition, discrete optimization methods such as graph-cuts and graph-shifts, and stochastic optimization methods such as data-driven Markov chain Monte Carlo. An emphasis will be placed on both the theoretical aspects of this field as well as the practical application of the models and inference algorithms.

Course Project: Each student will be required to implement a course project that is either a direct implementation of a method discussed during the semester or new research in Bayesian vision. A paper describing the project is required near the end of the semester (6-8 pages two column IEEE format). The papers will be peer-reviewed in the course; revisions need to be made based on the peer review and the final submission needs to include a letter to the editor describing the paper as if it is in submission to a journal and a description of the revisions made and why. Working project demos are required at the end of the semester. This is a ``projects'' course. Your projects can satisfy a Masters requirement. In most cases, it will involve at least some new/independent research. Previous offerings of this course have resulted in numerous papers accepted at major conferences and journals.

Prerequisites: It is assumed that the students have taken introductory courses in pattern recognition (CSE 555), and computer vision (CSE 573). Machine learning (CSE 574) is suggested but not required. A strong understanding and ability to work with probabilities, statistics, calculus and optimization is expected.

Permission of the instructor is required if these pre-requisites have not been met.

Course Goals: After taking the course, the student should will a clear understanding of the state-of-the-art models and inference algorithms for solving vision problems within a Bayesian methodology. Through completing the course project, the student will also have a deep understanding of the low-level details of a particular model/algorithm and application. The student will have completed some independent research in Bayesian Vision by the end of the course.

The student will also have experience in planning a project, conducting semi-independent research, and writing up the results; peer-review practice will also be part of the course.

Textbooks: There is unfortunately no complete textbook for this course. The required material will either be distributed by the instructor or found on reserve at the UB Library. Recommended textbooks are below; it is suggested you pick up a copy of at least one of the first three (and if all students do this there will be a half dozen copies of each floating around to share).

Li, S. Markov Random Field Modeling in Image Analysis. Springer-Verlag. 3rd Edition. 2009.
Winkler, G. Image Analysis, Random Fields and Markov Chain Monte Carlo Methods: A Mathematical Introduction. Springer. 2006.
Blake, A., Kohli, P. and Rother, C. Markov Random Fields for Vision and Image Processing. MIT Press. 2011.
Chalmond, B. Modeling and Inverse Problems in Image Analysis. Springer. 2003.
Koller, D. and Friedman, N. Probabilistic Graphical Models: Principles and Techniques. MIT Press. 2009.
Bishop, C. M. Pattern Recognition and Machine Learning. Springer. 2007.

Course Work

Grading: Letter grading distributed as follows:

In-Class Discussion/Quizzing (50%)
Homeworks (0%)
Project (50%)

In-Class Discussion/Quizzing: Half of the grades in this course are based on the students (1) participation in the class, (2) ability to answer questions when queried and (3) ask questions. No written quizzes are planned, but the professor reserves the possibility.

Homeworks: There will be weekly homeworks recommended. They will cover both theoretical and practical (implementation) aspects of the material. The homework assignments are not turned in. We will organize a weekly time where the students in the course will come together to discuss the weekly work without the professor around.

Programming Language: Student choice for the project (generally, Python, Matlab, Java, or C/C++). Any course-relevant aspects of the project need to be independently developed; e.g., if you are using belief propagation as your project's inference algorithm, then you need to implement belief propagation from scratch. No exceptions; don't ask.

For the homeworks and some in-class exercises we will use the UGM library written by Dr. Mark Schmidt; http://www.di.ens.fr/~mschmidt/Software/UGM.html. At various points in the course, you will be asked to either run through a demo/function from the library or implement/reimplement a different method for pedagogical value. There will be no introduction to the library in the course, you are expected to learn it in the first week or two (work through the early and simple demos ``Small,'' ``Chain,'' ``Tree'', and ``ICM.''

Word Processing: This course forces you to learn L^ATEX if you do not already know it. It is the language of the realm. All things submitted to me must be formatted in L^ATEX.

Project

The goal of the project is to have each student solve a real problem using the ideas learned herein. The professor will distribute/discuss project ideas in the first week of the class; students are encouraged to design their own project in conjunction with the professor. The ultimate goal is for each student to do some new work and learn by doing so. Within reason, camera and video equipment will be made available to the students from the VPML (my lab). Suitable arrangements should be made with the instructor to facilitate equipment use.

Project topics can cover a myriad of current problems in vision and must include some technical aspect developed on top of ideas in the course. A project focusing on statistics of a class of images/videos is also fair game but will need to be thoroughly justified.

Project Schedule

9/11

Project proposal due in class. 1-page description of the proposed project and the type of problem/data. It should include three milestones in planning. (All writing must be done in L^ATEX.)

9/20

Project plan due in class (this is the refinement of the project proposal; i.e., project proposal v 2). 3-page description of the proposed project, the most related work from the literature, the three milestones, planned data and experiments, and a goal statement that presents a table with two columns:

Outcome	Grade
My project will blah blah blah	A
My project will blah blah blah	B
My project will blah blah blah	C
My project will not work.	F

You fill in the blah blah blah and I'll consider it (and approve it or make you modify it). Hence, your Project plan is a contract and you have just graded yourself.

10/4

Milestone 1 Report due in class. (1-paragraph)

10/18

Milestone 2 Paper due in class. (4ish-pages)

11/1

Milestone 3 Paper due in class. (full paper)

11/1-6

Blind Peer Review Period. (Round robin with everyone reviewing two papers.)

11/15

Revised paper due. (Note 11/15 is the CVPR deadline.)

after 11/15

Project presentations and demos in class.

Project Write-Up

The paper should be in standard IEEE conference format at a maximum of 8 pages. We'll explain in class how to set it up. It should be approached as a standard paper containing introduction and related work, methodology, results, and discussion.

Working Course Outline

The course is roughly divided into two parts. In the first part, we discuss various modeling and associated learning algorithms. In the second part, we discuss the computing and inference algorithms which use the previously discussed models to solve complex inference problems in vision. The topic outline follows; citations are given and an underlined citation indicates a primary (must-read) one. All or most papers are available in PDF at the course directory (location above).

Paper citations are given below (somewhat sparsely), but few references are given to chapters in the books mentioned above. It is suggested you look in the books for more information when needed.

Introduction.
1. Discussion of Bayesian inference in the context of vision problems. (Winkler, 2006, Chapter 1) (Chalmond, 2003, Chapter 1) (Hanson, 1993) Probabilistic Inference Primer: (Griffiths and Yuille, 2006)
2. Presentation of relevant empirical findings concerning the statistics of images motivating the Bayesian approach. (Field, 1994) (Field, 1987) (Julesz, 1981) (Kersten, 1987) (Ruderman, 1994) (Simoncelli and Olshausen, 2001) (Torralba and Oliva, 2003) (Wu et al., 2007)
3. Model classes: discriminative, generative and descriptive. (Zhu, 2003)
Modeling and Learning.
1. Descriptive models on regular lattices.
  1. Markov random field models and Gibbs fields. (Li, 2001, §1.2) (Winkler, 2006, §2,3) (Dubes and Jain, 1989)
  2. The Hammersley-Clifford theorem.
  3. Bayes MRF Estimators (Winkler, 2006, §1.4) (Li, 2001, §1.5) (Geman and Geman, 1984)
  4. Examples:
    1. Auto-Models (Besag, 1974) (Li, 2001, §1.3.1, 2.3, 2.4) (Winkler, 2006, §15)
    2. Weak membrane models, Mumford-Shah, TV, etc.
  5. Applications:
    1. Image Restoration and Denoising (Li, 2001, §2.2)
    2. Edge Detection and Line Processes (Li, 2001, §2.3) (Geman and Geman, 1984)
    3. Texture (Li, 2001, §2.4) (Winkler, 2006, §15,16)
  6. MRF Parameter Estimation (Li, 2001, §6) (Winkler, 2006, §5,6)
    1. Maximum-Likelihood
    2. Pseudo-Likelihood
    3. Gibbs Sampler (and brief introduction to MCMC)
    4. Large Margin Methods (Blake et al., 2011, §15)
2. Descriptive Models on Regular Lattices: Advanced Topics
  1. Discontinuities and Smoothness Priors (Li, 2001, §4)
  2. FRAME and Minimax entropy learning of potential functionals. (Zhu et al., 1998) (Zhu et al., 1997) (Coughlan and Yuille, 2003)
  3. Hidden Markov random fields. (Zhang et al., 2001)
  4. Conditional random fields. (Lafferty et al., 2001) (Kumar and Hebert, 2003) (Wallach, 2004) (Ladicky et al., 2009)
  5. MRF as a foundation for multiresolution computing. (Gidas, 1989)
  6. Higher Order Extensions (Kohli et al., 2007) (Kohli et al., 2009) and Field of Experts (Roth and Black, 2009).
3. Descriptive and Generative Models on Irregular Graphs and Hierarchies.
  1. Markov random field hierarchies. (Derin and Elliott, 1987) (Krishnamachari and Chellappa, 1995) (Chardin and Perez, 1999)
  2. Over-Complete Bases and Sparse Coding (Zhu, 2003, §6) (Olshausen and Field, 1997) (Coifman and Wickerhauser, 1992)
  3. Textons (Julesz, 1981) (Zhu et al., 2005) (Malik et al., 1999)
  4. And-Or graphs and context-sensitive grammars. (Zhu and Mumford, 2007) (Han and Zhu, 2005)
  5. Dirichlet Processes (DP) and Bayesian Clustering (Ferguson, 1973)
  6. Latent Dirichlet Allocation, hierarchical DP and author-topic models. (Blei et al., 2003) (Teh et al., 2005) (Steyvers et al., 2004)
  7. Correspondence LDA (Blei and Jordan, 2003)
4. Integrating Descriptive and Generative Models (Guo et al., 2006)
Inference Algorithms.
1. Boundary methods.
  1. Level set evolution. (Chan and Vese, 2001)
  2. Region competition algorithm. (Zhu and Yuille, 1996a)
2. Exact Inference. Exploit the structure of the graph or the form of the potentials to search for the global optimum efficiently (in polynomial time).
  1. Chains and Trees.
  2. Sum-Product algorithm (exact Belief Propagation). (Bishop, 2006, §8) (Yedidia et al., 2001) (Frey and MacKay, 1997) (Felzenszwalb and Huttenlocher, 2006)
  3. Graph-Cuts: min-cut/max-flow relationship. (Blake et al., 2011, §2)
    What energy functions can/can not be minimized by graph cuts? (Kolmogorov and Zabih, 2004)
3. Approximate Inference.
  1. Discrete Deterministic Inference.
    1. Graph-Cuts: 1#1 -Expansion algorithm. (Boykov et al., 2001)
    2. Graph-Shifts algorithm. (Corso et al., 2007) (Corso et al., 2008b)
    3. Generalized Belief Propagation. (Yedidia et al., 2005) (Yedidia et al., 2000)
    4. Inference on And-Or graphs. (Zhu and Mumford, 2007) (Han and Zhu, 2005)
  2. Stochastic Inference. (Forsyth et al., 2001)
    1. Mean Field Approximation.
    2. Gibbs sampling. (Geman and Geman, 1984) (Winkler, 2006, §5,7)
    3. Metropolis-Hastings and Markov chain Monte Carlo methods. (Winkler, 2006, §10) (Tierney, 1994) (Liu, 2002)
    4. Data-Driven MarkovMCMC algorithm. (Tu and Zhu, 2002) (Tu et al., 2005) (Green, 1995)
    5. Swendsen-Wang algorithm. (Swendsen and Wang, 1987) (Barbu and Zhu, 2005) (Barbu and Zhu, 2004)
    6. Sequential MCMC and Particle Filters. (Isard and Blake, 1998) (Liu and Chen, 1998)

Additional Information

Similar Courses at Other Institutions: (incomplete and in no important order)

Professor Alan Yuille at UCLA. Vision as Bayesian Inference. http://www.stat.ucla.edu/~yuille/courses/Stat238/Stat_238.htm
Professor Song-Chun Zhu at UCLA. Statistical Modeling and Learning in Vision and Image Science. http://www.stat.ucla.edu/
Professor Song-Chun Zhu at UCLA. Statistical Computing and Inference in Vision and Image Science. http://www.stat.ucla.edu/
Professor Fei-Fei Li at Princeton. High-Level Recognition in Computer Vision http://vision.cs.princeton.edu/cs598_spring07/
Professor Tal Arbel at McGill. Statistical Computer Vision http://www.cim.mcgill.ca/~arbel/courses/626.html
Professor William T. Freeman at MIT. Advances in Computer Vision: Learning and Interfaces http://courses.csail.mit.edu/6.869/

Course Bibliography

Most items below have been cited above, but there are also some additional references that extend the content of the course. When available, PDFs of articles have been uploaded to the UBLearns ``Course Documents'' section. The naming convention is the first two characters of (up to) the first three authors following by an acronym for the venue (e.g., CVPR for Computer Vision and Pattern Recognition) followed by the year. So, the Geman and Geman 1984 PAMI article is GeGePAMI1984.pdf.

Bibliography

A. Barbu and S. C. Zhu.: Multigrid and Multi-level Swendsen-Wang Cuts for Hierarchic Graph Partitions.
In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 731-738, 2004.
A. Barbu and S. C. Zhu.: Generalizing Swendsen-Wang to Sampling Arbitrary Posterior Probabilities.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 27 (8): 1239-1253, 2005.
J. Besag.: Spatial interaction and the statistical analysis of lattice systems (with discussion).
J. Royal Stat. Soc., B, 36: 192-236, 1974.
J. Besag.: On the statistical analysis of dirty pictures (with discussion).
Journal of the Royal Statistical Society [Ser. B], 48: 259-302, 1986.
C. M. Bishop.: Pattern Recognition and Machine Learning.
Springer, 2006.
C. M. Bishop and J. M. Winn.: Non-linear Bayesian Image Modelling.
In European Conference on Computer Vision, volume 1, pages 3-17, 2000.
A. Blake, P. Kohli, and C. Rother, editors.: Markov Random Fields for Vision and Image Processing.
MIT Press, 2011.
D. M. Blei and M. I. Jordan.: Modeling Annotated Data.
In Proceedings of SIGIR, 2003.
D. M. Blei, A. Y. Ng, and M. I. Jordan.: Latent dirichlet allocation.
Journal of Machine Learning Research, 3: 993-1022, 2003.
C. A. Bouman and M. Shapiro.: A multiscale random field model for bayesian image segmentation.
Image Processing, IEEE Transactions on, 3 (2): 162-177, 1994.
Y. Boykov, O. Veksler, and R. Zabih.: Fast Approximate Energy Minimization via Graph Cuts.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 23 (11): 1222-1239, 2001.
B. Chalmond.: Modeling and Inverse Problems in Image Analysis, volume 155 of Applied Mathematical Sciences.
Springer, 2003.
T. F. Chan and L. A. Vese.: Active contours without edges.
IEEE Trans. on Image Processing, 10 (2): 266-277, 2001.
A. Chardin and P. Perez.: Semi-iterative inferences with hierarchical energy-based models for image analysis.
Energy Minimization Methods in Computer Vision and Pattern Recognition: Second International Workshop, EMMCVPR'99, York, UK, July 1999. Proceedings, pages 730-730, 1999.
URL http://www.springerlink.com/content/6yq1rglku6ccxjpu.
R. R. Coifman and M. V. Wickerhauser.: Entropy-based algorithms for best basis selection.
IEEE Transactions on Information Theory, 38 (2): 713-718, 1992.
T.F. Cootes and C.J. Taylor.: Statistical Models of Appearance for Computer Vision.
Technical report, Imaging Science and Biomedical Engineering, University of Manchester, 2004.
J. J. Corso, E. Sharon, and A. Yuille.: Multilevel Segmentation and Integrated Bayesian Model Classification with an Application to Brain Tumor Segmentation.
In Medical Image Computing and Computer Assisted Intervention, volume 2, pages 790-798, 2006.
J. J. Corso, Z. Tu, A. Yuille, and A. W. Toga.: Segmentation of Sub-Cortical Structures by the Graph-Shifts Algorithm.
In N. Karssemeijer and B. Lelieveldt, editors, Proceedings of Information Processing in Medical Imaging, pages 183-197, 2007.
J. J. Corso, E. Sharon, S. Dube, S. El-Saden, U. Sinha, and A. Yuille.: Efficient multilevel brain tumor segmentation with integrated bayesian model classification.
IEEE Transactions on Medical Imaging, 27 (5): 629-640, 2008a.
J. J. Corso, Z. Tu, and A. Yuille.: MRF Labeling with a Graph-Shifts Algorithm.
In Proceedings of International Workshop on Combinatorial Image Analysis, volume LNCS 4958, pages 172-184, 2008b.
J. M. Coughlan and A. L. Yuille.: Algorithms from Statistical Physics for Generative Models of Images.
Image and Vision Computing, Special Issue on Generative-Model Based Vision, 21 (1): 29-36, 2003.
H. Derin and H. Elliott.: Modeling and segmentation of noisy and texture images using gibbs random fields.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 9 (1): 39-55, 1987.
R. C. Dubes and A. K. Jain.: Random field models in image analysis.
Journal of Applied Statistics, 16 (2): 131 - 164, 1989.
L. Fei-Fei and P. Perona.: A Bayesian Hierarchical Model for Learning Natural Scene Categories.
In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2005.
P. F. Felzenszwalb and D. P. Huttenlocher.: Efficient Belief Propagation for Early Vision.
International Journal of Computer Vision, 70 (1), 2006.
T. S. Ferguson.: A bayesian analysis of some nonparametric problems.
The Annals of Statistics, 1 (2): 209-230, 1973.
D. J. Field.: Relations between the statistics of natural images and the response properties of cortical cells.
Journal of the Optical Society of America A, 4 (12): 2379-2394, 1987.
D. J. Field.: What is the goal of sensory coding?
Neural Computation, 6: 559-601, 1994.
D. Forsyth, J. Haddon, and S. Ioffe.: The joy of sampling.
International Journal of Computer Vision, 41 (1): 109-134, 2001.
B. J. Frey and D. MacKay.: A Revolution: Belief Propagation in Graphs with Cycles.
In Proceedings of Neural Information Processing Systems (NIPS), 1997.
S. Geman and D. Geman.: Stochastic Relaxation, Gibbs Distributions, and Bayesian Restoration of Images.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 6: 721-741, 1984.
B. Gidas.: A Renormalization Group Approach to Image Processing Problems.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 11 (2): 164-180, 1989.
ISSN 0162-8828.
doi: 10.1109/34.16712.
P. J. Green.: Reversible jump markov chain monte carlo computation and bayesian model determination.
Biometrika, 82 (4): 711-732, 1995.
T. L. Griffiths and A. Yuille.: Technical introduction: A primer on probabilistic inference.
Technical report, University of California at Los Angeles, 2006.
C. E. Guo, S. C. Zhu, and Y. N. Wu.: Modeling Visual Patterns by Integrating Descriptive and Generative Models.
International Journal of Computer Vision, 53 (1): 5-29, 2003.
C. E. Guo, S. C. Zhu, and Y. N. Wu.: Primal sketch: Integrating texture and structure.
Computer Vision and Image Understanding, 2006.
(to appear).
F. Han and S. C. Zhu.: Bottom-up/top-down image parsing by attribute graph grammar.
In Proceedings of International Conference on Computer Vision, volume 2, pages 1778-1785, 2005.
K. M. Hanson.: Introduction to Bayesian image analysis.
Medical Imaging: Image Processing, Proc. SPIE 1898: 716-731, 1993.
K. Held, E. R. Kops, B. J. Krause, III. Wells, W. M., R. Kikinis, and H. W. Muller-Gartner.: Markov random field segmentation of brain MR images.
Medical Imaging, IEEE Transactions on, 16 (6): 878-886, 1997.
M. Isard and A. Blake.: CONDENSATION - conditional density propagation for visual tracking.
International Journal of Computer Vision, 29 (1): 5-28, 1998.
B. Julesz.: Textons, the elements of texture perception and their interactions.
Nature, 290: 91-97, 1981.
D. Kersten.: Predictability and Redundancy of Natural Images.
Journal of the Optical Society of America, A 4 (12): 2395-2400, 1987.
P. Kohli, M. P. Kumar, and P. H. S. Torr.: P1#1 & beyond: Solving energies with higher order cliques.
In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2007.
P. Kohli, L. Ladicky, and P. H. S. Torr.: Robust higher order potentials for enforcing label consistency.
International Journal of Computer Vision, 82: 302-324, 2009.
V. Kolmogorov and R. Zabih.: What Energy Functions Can Be Minimized via Graph Cuts?
In European Conference on Computer Vision, volume 3, pages 65-81, 2002a.
V. Kolmogorov and R. Zabih.: What Energy Functions Can Be Minimized via Graph Cuts?
IEEE Transactions on Pattern Analysis and Machine Intelligence, 26 (2): 147-159, 2004.
Vladimir Kolmogorov and Ramin Zabih.: Multicamera Scene Reconstruction via Graph-Cuts.
In European Conference on Computer Vision, pages 82-96, 2002b.
S. Krishnamachari and R. Chellappa.: Multiresolution gmrf models for texture segmentation.
volume 4, pages 2407-2410 vol.4, 1995.
S. Kumar and M. Hebert.: Discriminative Random Fields: A Discriminative Framework for Contextual Interaction in Classification.
In International Conference on Computer Vision, 2003.
L. Ladicky, C. Russell, P. Kohli, and P. H. S. Torr.: Associative hierarchical crfs for object class image segmentation.
In Proceedings of International Conference on Computer Vision, 2009.
J. Lafferty, A. McCallum, and F. Pereira.: Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data.
In Proceedings of International Conference on Machine Learning, pages 282-289, 2001.
C. H. Lee, M. Schmidt, A. Murtha, A. Bistritz, J. Sander, and R. Greiner.: Segmenting brain tumor with conditional random fields and support vector machines.
In Proceedings of Workshop on Computer Vision for Biomedical Image Applications at International Conference on Computer Vision, pages 469-478, 2005.
S. Lee and M. M. Crawford.: Unsupervised multistage image classification using hierarchical clustering with a bayesian similarity measure.
Image Processing, IEEE Transactions on, 14 (3): 312-320, 2005.
S. Z. Li.: Markov Random Field Modeling in Image Analysis.
Springer-Verlag, 2nd edition, 2001.
J. S. Liu.: Monte Carlo Strategies in Scientific Computing.
Springer, 2002.
J. S. Liu and R. Chen.: Sequential monte carlo methods for dynamic systems.
Journal of the American Statistical Society, 93 (443): 1032-1044, 1998.
S. N. MacEachern and P. Muller.: Estimating mixture of dirichlet process models.
Journal of Computational and Graphical Statistics, 7 (2): 223-238, 1998.
J. Malik, S. Belongie, J. Shi, and T. Leung.: Textons, Contours, and Regions: Cue Combination in Image Segmentation.
In International Conference on Computer Vision, 1999.
M. R. Naphade and T. S. Huang.: A Probabilistic Framework for Semantic Video Indexing, Filtering, and Retrieval.
IEEE Transactions on Multimedia, 3 (1): 141-151, 2001.
B. A. Olshausen and D. J. Field.: Sparse coding with an overcomplete basis set: A strategy employed by v1?
Vision Research, 37 (23): 3311-3325, 1997.
A. Raj and R. Zabih.: A graph cut algorithm for generalized image deconvolution.
In Proceedings of International Conference on Computer Vision, 2005.
A. Ranganathan.: The dirichlet process mixture (dpm) model.
September 2004.
URL http://www.cs.rochester.edu/~michalak/mlseminar/fall05/dirichlet.pdf.
S. Richardson and P. J. Green.: On Bayesian Analysis of Mixtures With an Unknown Number of Components.
Journal of the Royal Statistical Society - Series B, 59 (4): 731-758, 1997.
S. Roth and M. J. Black.: Fields of experts.
International Journal of Computer Vision, 82 (2): 205-229, 2009.
D. L. Ruderman.: The statistics of natural images.
Network: Computation in Neural Systems, 5 (4): 517-548, 1994.
M. Schaap, I. Smal, C. Metz, T. van Walsum, and W. Niessen.: Bayesian Tracking of Elongated Structures in 3D Images.
In N. Karssemeijer and B. Lelieveldt, editors, Proceedings of Information Processing in Medical Imaging, 2007.
E. P. Simoncelli and B. A. Olshausen.: Natural image statistics and neural representation.
Annual Review of Neuroscience, 24: 1193-1216, 2001.
M. Steyvers, P. Smyth, M. Rosen-Zvi, and T. Griffiths.: Probabilistic author-topic models for information discovery.
In 10th ACM SigKDD Conference on Knowledge Discovery and Data Mining, 2004.
E. B. Sudderth, A. Torralba, W. T. Freeman, and A. S. Willsky.: Describing visual scenes using transformed dirichlet processes.
In Proceedings of Neural Information Processing Systems (NIPS), 2005.
R. H. Swendsen and J. S. Wang.: Nonuniversal Critical Dynamics in Monte Carlo Simulations.
Physical Review Letters, 58 (2): 86-88, 1987.
Y. W. Teh, M. I. Jordan, M. J. Beal, and D. M. Blei.: Hierarchical dirichlet processes.
In Advances in Neural Information Processing Systems (NIPS) 17, 2005.
L. Tierney.: Markov chains for exploring posterior distributions.
The Annals of Statistics, 22 (4): 1701-1728, 1994.
Phil Torr and C. Davidson.: IMPSAC: Synthesis of Importance Sampling and Random Sample Consensus.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 25 (3): 354-364, 2003.
A. Torralba and A. Oliva.: Statistics of natural image categories.
Network: Computation in Neural Systems, 14: 391-412, 2003.
F. Torre and M. J. Black.: Robust Principal Component Analysis for Computer Vision.
In International Conference on Computer Vision, 2001.
Z. Tu and S. C. Zhu.: Image Segmentation by Data-Driven Markov Chain Monte Carlo.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 24 (5): 657-673, 2002.
Z. Tu, X. R. Chen, A. L. Yuille, and S. C. Zhu.: Image Parsing: Unifying Segmentation, Detection and Recognition.
International Journal of Computer Vision, 63 (2): 113-140, 2005.
H. M. Wallach.: Conditional Random Fields: An Introduction.
CIS MS-CIS-04-21, University of Pennsylvania, 2004.
G. Winkler.: Image Analysis, Random Fields, and Markov Chain Monte Carlo Methods.
Springer, 2nd edition, 2006.
Y. N. Wu, S. C. Zhu, and C. E. Guo.: From Information Scaling of Natural Images to Regimes of Statistical Models.
Quarterly of Applied Mathematics, 2007.
J. S. Yedidia, W. T. Freeman, and Y. Weiss.: Generalized belief propagation.
In Advances in Neural Information Processing Systems (NIPS), volume 13, pages 689-695, 2000.
J. S. Yedidia, W. T. Freeman, and Y. Weiss.: Bethe free energy, Kikuchi approximations and belief propagation algorithms.
Technical Report TR2001-16, Mitsubishi Electronic Research Laboratories, May 2001.
J. S. Yedidia, W. T. Freeman, and Y. Weiss.: Constructing Free-Energy Approximations and Generalized Belief Propagation Algorithms.
IEEE Transactions on Information Theory, 51 (7): 2282-2312, 2005.
Ramin Zabih and Vladimir Kolmogorov.: Spatially Coherent Clustering Using Graph Cuts.
In IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 437-444, 2004.
Y. Zhang, M. Brady, and S. Smith.: Segmentation of brain mr images through a hidden markov random field model and the expectation-maximization algorithm.
IEEE Transactions on Medical Imaging, 20 (1): 45-57, January 2001.
S. C. Zhu.: Stochastic jump-diffusion process for computing medial axes in markov random fields.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 21 (11): 1158-1169, 1999.
S. C. Zhu.: Statistical Modeling and Conceptualization of Visual Patterns.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 25 (6): 691-712, 2003.
S. C. Zhu and D. Mumford.: Prior learning and gibbs reaction-diffusion.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 19 (11): 1236-1250, 1997.
S. C. Zhu and D. Mumford.: A stochastic grammar of images.
Foundations and Trends in Computer Graphics and Vision, 2 (4): 259-362, 2007.
S. C. Zhu and A. Yuille.: Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 18 (9): 884-900, 1996a.
S. C. Zhu and Alan L. Yuille.: FORMS: A Flexible Object Recognition and Modeling System.
International Journal of Computer Vision, 20 (3): 187-212, 1996b.
S. C. Zhu, Y. Wu, and D. Mumford.: Minimax entropy principle and its application to texture modeling.
Neural Computation, 9 (8): 1627-1660, 1997.
S. C. Zhu, Y. N. Wu, and D. B. Mumford.: FRAME: Filters, Random field And Maximum Entropy: -- Towards a Unified Theory for Texture Modeling.
International Journal of Computer Vision, 27 (2): 1-20, 1998.
S. C. Zhu, C. E. Guo, Y. Wang, and Z. Xu.: What are textons?
International Journal of Computer Vision, 62 (1): 121-143, 2005.