Activity recognition is one of the most active topics within computer
vision. Despite its popularity, its application in real life scenarios is
limited because many methods are not entirely automated and consume high
computational resources for inferring information. In this work, we
contribute two novel algorithms: (a) one for automatic video sequence
segmentation - elsewhere referred to as action spotting or action detection
- and (b) a second one for reducing activity representation computational
cost. Two Bag-of-Words (BoW) representation schemas were tested for
recognition purposes. A set of experiments was performed, both on publicly
available datasets of activities of daily living (ADL), but also on our own
ADL dataset with both healthy subjects and people with dementia, in
realistic, life-like environments that are more challenging than those of
benchmark datasets. Our method is shown to provide results better than, or
comparable with, the SoA, while we also contribute a realistic ADL dataset
to the community.
Recognition of Activities of Daily Living
22.07.2013
Ημερομηνία : 22.07.2013
Ώρα : 11:00-12:30
Μέρος : Aίθουσα Συναντήσεων "Μεσογειακών Σπουδών", ΙΤΕ, Ηράκλειο, Κρήτη
Φιλοξενείται από : Τσακαλίδης Παναγιώτης
Alexia Briassouli received the Diploma degree in eletrical engineering from
the National Technical University of Athens in 1999, an MSc in Signals and
Systems Processing from the University of Patras in 2000, and the Ph.D.
degree from the Department of Electrical and Computer Engineering at the
University of Illinois in Urbana Champaign in 2005. She taught signal,
image, video and audio processing classes at an undergraduate and graduate
level as an adjunct professor in the Department of Computer and
Communication Engineering at the University of Thessaly in Volos, Greece
from 2006 to 2010 and supervised over 10 diploma theses. She has been
working as a postdoctoral research fellow at CERTH since 2006, where she is
co-supervising 2 PhD's, participating in the writing of research proposals
and carrying out both independent research and research within EU projects.
Her current research interests include statistical image and video
processing, crowd motion analysis and event detection, human activity
recognition. She has authored over 46 publications in peer-reviewed
journals, conferences and books, including IEEE Trans. on Pattern Analysis
for Machine Intelligence, IEEE Trans. on Image Processing, IEEE Trans. on
Circuits and Systems for Video Technology, IEEE Trans. on Multimedia, ECCV,
ACCV, ICCV, ICPR, ICIP, and has participated in a number of European and
National projects. She was involved in video analysis for activity
recognition in the FP6 project AceMedia; event detection in IST STREP JUMAS;
and fire, smoke and event detection in the EDA project MEDUSA. Currently she
is working on event detection and abnormal activity detection in crowd
videos for the FP7 project Dem@Care, for remote monitoring and care of
people with dementia, to help them continue living independently at home.