ICCVG 2016
monday, 25 September 2017
switch to english

Sign in

Registered user

Login

Password

Home > Description

Invited Lectures
Professor Amitava Datta

The University of Western Australia
School of Computer Science & Software Engineering

www.web.uwa.edu.au/people/amitava.datta
 

Professor Wojciech Chojnacki

University of Adelaide
School of Computer Science

cs.adelaide.edu.au/~wojtek/
 

Dr Ika Rogowska

The University of Utah
Departament of Psychiatry
Brain Institute


 

Lecture title:
SUBSCALE: an efficient, scalable and parallelizable algorithm for subspace clustering

Summary

Clustering is one of the most important unsupervised learning algorithms. It is the process of automatically finding groups of similar data points in the space of the dimensions or attributes of a dataset. Data group together differently under different subsets of dimensions, called subspaces. Quite often a dataset can be better understood by clustering it in its subspaces, a process called subspace clustering. But the exponential growth in the number of these subspaces with the dimensionality of data makes the whole process of subspace clustering computationally very expensive. There is a growing demand for efficient and scalable subspace clustering algorithms in many application domains like computer vision, astronomy, biology and social networking. The performance of the existing algorithms deteriorates drastically with the increase in the number of dimensions. Most of these algorithms require multiple data scans and generate a large number of redundant subspace clusters, either implicitly or explicitly, during the clustering process. In this talk, we present SUBSCALE, a novel clustering algorithm to find non-trivial subspace clusters with minimal cost. Our algorithm scales very well with the dimensionality of the dataset and is highly parallelizable. We will give examples of evaluation of our algorithm on computer vision datasets, both for sequential and multicore executions.

Lecture title:
Recent advances in ellipse estimation techniques

Summary

The task of fitting an ellipse to data is frequently encountered in numerous disciplines, including computer vision and image processing. This is partially explained by the fact that even though very few shapes and trajectories are perfectly elliptical, non-elliptical shapes and trajectories can often be modelled using one or more ellipses. The talk will present various parameter estimation methods for ellipse fitting. A special focus will be given to recent ellipse-specific methods guaranteeing that estimates take the form of ellipses, not just conics.

Lecture title:
Resting State Connectivity in Mild Traumatic Brain Injury Using MRI

Summary

Functional magnetic resonance imaging (fMRI) measures the metabolic changes that occur within the brain. It may be used to examine the brain’s anatomy, determine which parts of the brain are handling critical functions, evaluate the effects of stroke or disease, or guide brain treatments. In particular, we can use resting state fMRI, which looks at brain connectivity between regions that are functionally linked, to detect abnormalities within the brain that cannot be found with other imaging techniques. One such abnormality of interest is mild traumatic brain injury (mTBI), which affects millions of people/year globally. mTBI is associated with a host of clinical, neurobiological and psychosocial factors, which have not been fully elucidated, including cognitive and neuropsychiatric disorders. Given the limited sensitivity of standard clinical imaging approaches for detecting neurobiological changes in mTBI, we applied the resting state fMRI technique and investigated the functional networks in individuals with mTBI and in a group of healthy comparison subjects.

back           
Asociation for Image Processing Faculty of Applied Informatics and Mathematics, Warsaw University of Life Sciences Polish-Japanese Academy of Information Technology Faculty of Information Science, West Pomeranian University of Technology Springer, Lecture Notes in Computer Science

Association for Image Processing

Faculty of Applied Informatics and Mathematics

Photo of Warsaw by www.zdjeciawarszawy.pl

v. 2012.5.1.1 eConf © 2008-2016 Piotr Kužniacki

Last modification 24-09-2016

Site counter 13811 Online users 14