Analysis of Daily Activity for Long-Term Care of People with Special Needs

Speaker: Marilyn Wolf
Affiliation: Georgia Institute of Technology

Abstract:  Many people with cognitive and physical disabilities require some amount of assistance in daily living for their entire lives. Assistance ranges from mothering-style assistance to physical assistance with the most basic daily necessities. This population is underserved; much care is provided by a combination of family members and untrained staff. IoT systems can help provide both localized data and big-picture analysis to help these caregivers use their time effectively.  This talk will discuss several results based on collaborations with my student Honggab Kim as well as Mihaela van der Schaar and Cathy Bodine.  These results center on movement around the house: records of movement can be used to provide required logs of activity; comparison of day-to-day activity may also be useful as an indicator of wellness.  Multitarget tracking uses observations from cameras distributed around the house, as well as other types of sensors that detect activity but no subject identification.  Improvements in the problem, formulation and algorithms aid in the generation of tracks from observations. Adding non-appearance-bearing sensor observations can be used to generate tracks with modest loss in accuracy. Clustering analysis can be used to identify typical and atypical activity patterns. Both Levenshtein edit distance and dynamic time warping have been used in other applications but introduce some limitations; we will discuss the use of dynamic event warping as a metric for track clustering.

Biography:  Marilyn Wolf is Rhesa S. “Ray” Farmer Distinguished Chair in Embedded Computing Systems and Georgia Research Alliance Eminent Scholar at the Georgia Institute of Technology.  She received her BS, MS, and PhD in electrical engineering from Stanford University in 1980, 1981, and 1984, respectively.  She was with AT&T Bell Laboratories from 1984 to 1989.  She was on the faculty of Princeton University from 1989 to 2007.  Her research interests include cyber-physical systems, embedded computing, embedded video and computer vision, and VLSI systems. She has received the ASEE Terman Award and IEEE Circuits and Systems Society Education Award. She is a Fellow of the IEEE and ACM and an IEEE Computer Society Golden Core member.

For more information, contact Prof. Mani Srivastava ()

Date(s) - Oct 10, 2017
2:00 pm - 3:30 pm

E-IV Tesla Room #53-125
420 Westwood Plaza - 5th Flr., Los Angeles CA 90095