Machine Learning Approaches to Face Image Analysis and Image Classification

Date: 
26.4.2016 13:00

 

Infotech Oulu Lecture Series

Lecturer: Professor Fadi Dornaika, University of the Basque Country, Spain

Duration:  Two presentations of about one hour each.

Date: Tuesday, April 26, 2016
Time: 13:00-15:00
Room: TS128

Summary

The first part of the talk will present some of our research on face gesture tracking and expression recognition using still images and videos. The topics that will be addressed are:
1) Model-based face and facial gesture tracking using Online Appearance Models.
2) Dynamic facial expression recognition.
3) Simultaneous tracking and recognition.
4) Machine learning (model-less) 3D face pose estimation.

The second part will present some machine learning paradigms for image classification/face recognition. The topics that will be discussed include:

1) Feature selection in embedded sub-spaces
2) Cooperative feature selection-embedding
3) Supervised Laplacian Eigenmaps
4) Exponential Linear Discriminant Embedding
5) Graph-based semi-supervised embedding for pattern classification

Short Bio

Fadi Dornaika received his Master degree in signal, image and speech processing from Grenoble Institute of Technology, France, in 1992, and a Ph.D. degree in computer science from Grenoble Institute of Technology and INRIA, in 1995.  Currently, he is a Research Professor at IKERBASQUE (Basque Foundation for Science) and the University of the Basque Country (UPV/EHU). Prior to joining IKERBASQUE, he held numerous research positions at several research centers and universities in Europe, China, and Canada. His research covers a broad spectrum in computer vision and pattern recognition. He has published more than 190 papers in the field of computer vision and pattern recognition.  His current research includes pattern recognition, machine learning and data mining.

More information: Abdenour Hadid

Last updated: 5.4.2016