Home Tutorials

The following Professors are offering tutorials at MCPR on June 29, 2013.

Professor Raúl Rojas


Free University of Berlin



Hybrid classifiers

- Pattern recognition
- Nearest Neighbours
- AdaBoost for Face Recognition
- Random Forests

Raúl Rojas webpage

Professor Robert Pless


Washington University in St. Louis


Robert Pless is a Professor of Computer Science and Engineering at Washington University in St. Louis. His research focus is data driven approach to understanding motion and change in video, with a current focus on long term time-lapse imagery. Dr. Pless has a Bachelors Degree in Computer Science from Cornell University in 1994 and a PhD from the University of Maryland, College Park in 2000. He received the NSF CAREER award in 2006.  He has served as chair of the IEEE Workshop on Omnidirectional Vision and Camera Networks (OMNIVIS) in 2003, the MICCAI Workshop on Manifold Learning in Medical Imagery in 2008, and the IEEE Workshop on Motion and Video Computing in 2009.


Robert Pless webpage


Tutorial Description:


Linear Image Decompositions through Time: More than Dimensionality Reduction



Linear methods such as Principal Component Analysis, Canonical Components Analysis are a cornerstone of image analysis methods.  In this tutorial, I will provide an introduction to classical and novel component analysis techniques, with a focus on how can be learned by applying these methods to video or time-lapse data.

The first part of the tutorial will review traditional linear techniques such as PCA, LDA, CCA, etc, and will explore extensions that support incremental solutions, often important to apply these methods to large image data sets.  When applied to time-varying imagery, these methods create image components and coefficients that have substantial temporal structure.  The tutorial will discuss methods of visualizing this structure, and tools that build on this structure to give novel approaches to scene segmentation and understanding motion in video.



Linear Models

  • Review of PCA/SVD, LDA, CCA,
  • Robust principal component analysis,
  • Principal component analysis with uncertainty/missing data,
  • Incremental PCA.


  • Data visualization,
  • Temporal decompositions,
  • Applications to motion analysis and segmentation.

Professor Roberto Manduchi


Departament of Computer Engineering
University of California at Santa Cruz, USA


On Image Noise, or: How To Take a Well Exposed Picture (or a Stack Thereof)


Robert Manduchi webpage


Professor Edgar F. Roman-Rangel


Research Assistant @ Viper group
Computer Vision and Multimedia Lab
CUI - University of Geneva




Edgar Roman-Rangel is a research assistant at the Computer Vision and Multimedia Laboratory of the University of Geneva. His research interests include Computer Vision, Information Retrieval, and Machine Learning, specially applied to real world needs such as cultural heritage, bioinformatics, and social develpment. He obtained his Master degree in Computer Science from the Instituto Tecnológico y de Estudios Superiores de Monterrey (ITESM) in 2006, and his PhD from the École Politechnique Fédérale de Lausanne (EPFL) in 2013.



Review of shape representations



In this tutorial I will introduce several techniques for description of shapes, providing insights about their advantages and drawbacks. The tutorial will continue with the introduction of the projects CODICES and MAAYA that explore shape representations for Maya hieroglyphs, and that have been conducted at the Idiap research institute. Finally, the tutorial will conclude with a short lab that will provide the opportunity to experiment with some of the methods that were presented.



Review of shape descriptors (1h30’)

  • Fourier and moments
  • Matching techniques
  • SIFT and HOG
  • The network of adjacent segments kAS
  • Shape Context descriptors

A real example (30’)

  • The CODICES project at Idiap

Optional Hands-on lab (1h)

  • Exercise to compare the performance of selected descriptors retrieving shapes from the MPEG-7 Core Experiment CE-Shape-1 dataset (Matlab code will be provided).
  • It is recommended to bring a laptop with Matlab.

Edgar F. webpage


Professor Sai Ravela


Sai Ravela

Principal Research Scientist, EAPS(PAOC) @ MIT.



Alignment Models in Vision

Sai Ravela webpage





Who is online?

We have 2 guests online

Administrated by CICATA Querétaro IPN