CS Colloquium Series @ UCY
Department of Computer Science - University of Cyprus
The Department of Computer Science at the University of Cyprus holds research colloquiums and social hours approximately once weekly. All university students, faculty, and staff are invited to attend. Notifications about new and upcoming events are automatically disseminated to a variety of institutional lists.
If you don't receive these notifications, but want to get informed about upcoming colloquium announcements, you can do the following:
List
RSS
Directions
Colloquium Coordinator: Demetris Zeinalipour
Colloquium: Real-Time Hand Tracking for Human-Computer Interaction, Dr. Paris Kaimakis (University of Cyprus, Cyprus), Wednesday, December 18, 2013, 10:00-11:00 EET.
The Department of Computer Science at the University of Cyprus cordially invites you to the Colloquium entitled:
Real-Time Hand Tracking for Human-Computer Interaction
 |
Speaker: Dr. Paris Kaimakis |
Abstract:
The successful commercialisation of Microsoft's Kinect, and the recent
release of the Leap Motion Controller have demonstrated the increasing
demand for reliable, low-cost, real-time Markerless Motion Capture
systems -- systems that can sequentially estimate the subject’s 3Dpose based on visual information, without markers. Such systems have
numerous lucrative applications, e.g. in human-computer interaction,
computer gaming, animation, automated surveillance, and more.
The last decade has seen limited progress in the field, mainly due to
the conflicting design criteria of such systems: real-time processing
indicates that implementation of gradient-based algorithms is
necessary for tracking, while the low-cost criterion disqualifies the
use of dedicated hardware and therefore imposes use of low-level and
often ambiguous data. The problem is that gradient-based approaches
become very unstable in the presence of such ambiguity.
In this talk we introduce a system designed to circumvent these
difficulties, achieving real-time tracking of the human hand with high
precision and zero hardware requirements (further to an ordinary
laptop). The talk will focus on the system's early design and
modelling steps, as well as its Bayesian formulation.
Short Bio:
Dr Paris Kaimakis conducted his doctoral research in the Signal
Processing Laboratory at the University of Cambridge focusing on
solutions to the Markerless Motion Capture problem. After receiving
the PhD degree in 2009, he worked as a Research Associate in the
Computer Laboratory, again in the University of Cambridge. His
post-doctoral research concentrated on the automation of urban road
traffic surveillance and flow-volume estimation using visual means and
sequential Monte Carlo methods. In 2010 he joined Autonomy, the UK's
largest software firm, as a Research and Development Engineer. While
in Autonomy, he worked in the development of augmented reality
applications for the iPhone/iPad and for Android devices. He has been
involved in the development of the popular Aurasma App, the world's
first visual browser (over 3 million users, available at the App Store
and Google Play) and worked in related side-projects such as 3D head
tracking, image processing and image analysis. In 2011 Paris returned
to Cyprus and co-founded a company dedicated to the development and
commercialisation of new-generation, human-oriented, controller-free
computer interfaces. Paris has also conducted EU-funded research on
the automation of visual surveillance on Cypriot sea borders, and has
taught computer vision to undergraduates at the University of Cyprus.
Sponsor: The CS Colloquium Series is supported by a generous donation from  |