Some Tutorials

“Object Registration and Tracking from a Learning Perspective” [video of the lecture] (March 2008)

I was invited to teach at the Machine Learning Summer School (MLSS’08) in March 2008 graciously hosted by NICTA/ANU in Kioloa, Australia. 

Abstract:- In his tutorial I cover some of the core fundamentals in vision and demonstrate how they can be interpreted in terms of machine learning fundamentals. Unbeknownst to most researchers in the field of machine learning, the fundamentals of object registration and tracking such as optical flow, interest descriptors (e.g., SIFT), segmentation and correlation filters are inherently related to the learning topics of regression, regularization, graphical models, generative models and discriminative models. As a result many aspects of vision can be interpreted as applied forms of learning. From this discussion on fundamentals we shall also explore advanced topics in object registration and tracking such as non-rigid object alignment/ tracking and non-rigid structure from motion and how the application of machine learning is continuing to improve these technologies.

  1. Lecture 1 [pdf]

  2. Lecture 2 [pdf]

  3. Lecture 3 [pdf]

  4. Lecture 4 [pdf]

  5. Lecture 5 [pdf]

“An Introduction to Speaker Verification” (September 2002)

I was invited to give a 2 day tutorial on speaker verification to the Sony Company in San Jose in September 2002. In this tutorial we cover speech production, cepstral coefficients, GMMs, background models, etc.

  1. Lecture [ppt]

Some Invited Talks

“Discriminative Convex Models for Non-Rigid Object Alignment” [pdf] variants of talk given at CMU (August 2008), UBC (June 2008) and ANU (March 2008).

“ Patch-Based Face Analysis for Viewpoint Invariant Face Recognition” [pdf] variants of talk given at CMU (August 2007), IDIAP (July 2007).

AdHoc Tutorials

“Multilinear Analysis (MLA) - Tensorface Tutorial” [ppt] brief tutorial on multilinear analysis and tensor faces.

“Introduction to Sparse Learning” [pdf] brief tutorial on sparse learning through maximum margin and lasso regression.