Lectures

Date	Lecture	Reading
Wednesday 09/21	Lecture 1: Introduction Lecture notes: slides
Wednesday 09/28	Lecture 2: Dimensionality reduction (I) Principal component analysis, factor analysis, probabilistic principal component analysis. Lecture notes: slides	Required: Tenenbaum et al.: Isomap Roweis and Saul: LLE Further study: Eigenface Visual learning and recognition
Thursday 09/29	Lecture 3: Dimensionality reduction (II) mixture of Gaussians, EM algorithm, mixture of probabilistic principal component analyzers, mixture of factor analyzers Lecture notes: slides Supplementary reading: Zoubin Ghahramai’s lecture notes on the EM algorithm Chris Bishop’s lectures notes on Mixture of Gaussians	Required: Roweis et al.: Global coordination Teh and Roweis: Alignment of local representation Further study: Factor analysis and mixture of factor analyzers Probabilistic PCA Mixture of PPCA Laplacian eigenmap Spectral clustering Unified view of spectral embedding and clustering
Monday 10/03	Lecture 4: Dimensionality reduction (III) EM algorithm, nonlinear dimensionality reduction, isometric mapping, locally linear embedding.. Lecture notes: slides	Required: Belhumeur et al.: Fisherfaces vs. Eigenfaces Pontil and Verri: SVM for object recognition Mohan et al.: SVM-based object detection Further study: RVM applications to 3D pose estimation RVM application to visual tracking
Tuesday 10/04	Lecture 5: Classifier (I) Linear regression, logistic regression, Fisher linear discriminate Lecture notes: slides Supplementary reading: Tommi Jaakkolla lecture notes: 3, 4, 5, 6.
Wednesday 10/19	Lecture 6: Classifier (II) Support vector machines, relevance vector machine Lecture notes: slides Supplementary reading: Chris Burges’ SVM tutorial Bernhard Scholkopf SVM and kernel methods Bernhard Scholkopf Support vector learning Resources: http://www.kernel-machines.org/	Required (due Nov 1): Agarwal and Triggs: 3D human pose estimation using RVM Further study: Viola et al: Adaboost-based real-time pedestrian detection Tipping: Relevance vector machine Williams et al: Visual tracking using RVM Avidan: Support vector tracking Romdhani et al: Efficient SVM-based face detector
Thursday 10/20	Lecture 7: Classifier (III) Kernel principal component analysis, kernel discriminant analysis, Bagging, Adaboost, and applications Lecture notes: slides Supplementary reading: Robert Schapire: A boosting tutorial Resources: http://www.boosting.org/	Required (due Nov 8): Viola and Jones: Real-time Adaboost-based face detector Viola et al: Adaboost-based real-time pedestrian detection Avidan: Ensemble tracking Further study: Scholkopf: Kernel PCA Mika et al: Kernel Fisher discriminant Bauer and Kohavi: Empirical comparisons of Bagging, Boosting and variants.
Monday 10/24	Lecture 8: Graphical model (I) Introduction Lecture notes: slides Supplementary reading: Kevin Murphy: A brief introduction to graphical models and Bayesian Networks Zoubin Ghahramani: Unsupervised learning lecture notes Brendan Frey: CVPR 00 tutorial Blake, Freeman, Bishop and Viola: ICCV 03 tutorial Christopher Bishop: ECCV 04 tutorial Resources: Intel Open Source Probabilistic Network Library Kevin Murphy's Bayes Net Toolbox for Matlab Microsoft Bayesian Network Editor and Toolkit	Required (due Nov 15): Tipping: Relevance vector machine
Tuesday 10/25	Lecture 9: Graphical model (II) EM algorithm, sampling algorithms, MCMC methods Lecture notes: slides Supplementary reading: David MacKay: Introduction to MCMC methods Christopher Bishop: BCS lecture notes	Required (due Nov 29): Forsyth et al.: The joy of sampling Tu et al.: Image parsing Further study: Zhu et al: Filter, random fields, and maximum entropy (FRAME)
Monday 11/21	Lecture 10: Graphical model (III) Markov chains, MCMC methods and applications Lecture notes: slides Supplementary reading: Zhu et al.: ICCV MCMC tutorial
Tuesday 11/22	Lecture 11: Graphical model (IV) Belief propagation, variational inference Lecture notes: slides DDBPMC for human pose estimation Supplementary reading: Jordan et al.: Introduction to variational methods Yedidia: Understanding belief propagation	Required (due Dec 6): Freeman et al.: Learning low level vision
Wednesday 11/23	Lecture 12: Project midterm presentation
Thursday 11/24	Lecture 13: Visual tracking (I) Condensation algorithm and applications Lecture notes: slides	Required (due Dec 13): Isard and Blake: Condensation algorithm Toyama and Blake: Visual tracking with Condensation and exemplars Further study: Lim et al.: Incremental learning for visual tracking Black and Jepson: Eigentracking
Monday 12/26	Lecture 14: Visual Tracking (II) Variants of particle filters, WSL model, 3D human tracking Lecture notes: slides	Further study: Black and Jepson: Eigentracking Jepson et al: WSL model
Tuesday 12/27	Lecture 15: Visual Tracking (III), Dynamics Non-parametric density estimation, adaptive visual tracking, mean shift algorithm and applications. Lecture notes: slides Incremental learning for robust tracking	Further study: Comaniciu and Meer: Mean shift application Comaniciu et al.: Mean shift tracking
Wednesday 12/28	Lecture 16: Term project presentation Term projects
Thursday 12/29	Lecture 17: Tem project presentations/wrap up Lecture notes: slides