ICVSS 2009 - International Computer Vision Summer School 2009

Machine Learning for Computer Vision

Home

School Committee

School Courses

Deadlines

Application

Registration

Location and Accommodation

Sponsors

Call for Student Posters

Poster Submission

Best Presentation Prize

Scholarship

Local Arrangements

Previous Editions

Detailed ICVSS 2009 Programme

Lectures

Speakers

Syllabus

Titles & Abstracts

Michael Black
Brown University, USA

Review of traditional 3D body tracking,
3D bodies from 2D pictures, body shape estimation.

Human body shape and pose from images

Andrew Fitzgibbon
Microsoft Research Ltd, Cambridge, UK

Bayes, priors, MAP estimation, Markov random fields, image-based rendering, new-view synthesis, stereo, shape from silhouette

Seeing what you want to see: Bayes and priors in vision problems

David Forsyth
University of Illinois at
Urbana-Champaign, USA

Methods for reasoning about relations between words and pictures.

Words and pictures

Zoubin Ghahramani
University of Cambridge, UK

Probabilistic approaches, Bayesian machine learning and AI, approximation and nonparametrics algorithms.

Probabilistic approaches to Machine Learning

Dan Huttenlocher
Cornell University, USA

Object Category Recognition, Relational Models, Scene Context, Distance Transforms, Viterbi Algorithm, Generative and Discriminative Learning.

Pictorial structure models for object category recognition

Takeo Kanade
Carnegie Mellon University, USA

Computer Vision and Machine Learning: the past, the present, and the future

Computer Vision and Machine Learning: the past, the present and the future

Stefano Soatto
UCLA, USA

Visual representation, learning, invariance, sufficient statistics, entropy, coding length.

Where is the “Information” in an image, and can it be learned?

Advanced Tutorials

Speakers

Syllabus

Titles & Abstracts

Nello Cristianini
University of Bristol, UK

Linear Pattern Functions, Non-linear Patterns via kernel functions,
Statistical learning theory,
Optimization Theory, Various kernel based algorithms, Various kernel functions, Examples.

Kernel methods for Pattern Analysis

Rob Fergus
New York University, USA

Large scale computer vision,
Internet vision,
Distance metric learning,
Hashing techniques.

Machine learning methods for large-scale image search and recognition

Pushmeet Kohli
Microsoft Research Ltd, Cambridge, UK

St-Mincut, Maxflow, Submodular functions, Labelling Problems, Image Segmentation, Object Recognition.

Graph Cuts for energy minimization

Tae-Kyun Kim
University of Cambridge, UK

Boosting, mixture of experts, multiple classifier learning, Random forest, locally linear models, object recognition, tracking, class segmentation

Learning multiple classifiers in computer vision

John Winn
Microsoft Research Ltd, Cambridge, UK

Probabilistic models, graphical models,
generative vs. discriminative models,
Bayesian inference.

Probabilistic models for
understanding images

Reading Group - with prize!
The prize was assigned to:
Mihoko Shimano, University of Tokyo, Japan

Speakers

Titles, Abstracts, Homework, Syllabus

Stefano Soatto
UCLA, USA

A critical review of old ideas, with treasure hunt

Industrial and Demo Session

Speakers

Syllabus

Titles & Abstracts

Dariu Gavrila
Daimler Research and University of Amsterdam

Intelligent vehicles, pedestrian detection, benchmarking, multi-cue object recognition.

Intelligent vehicles that (learn to) see

Matthew Johnson
Nokia Point and Find
San Francisco, USA

Real-time computer vision, object recognition, ubiquitous computing, embedded devices

Real-time Computer Vision on embedded devices

Björn Stenger
Toshiba Research Europe Ltd,
Cambridge, UK

Multi-cue Tracking, HCI

Visual Remote Control

For more information, send an email to: icvss@dmi.unict.it