I am an Assistant Professor in the Department of Electrical and Computer Engineering at The University of Texas at Austin, where I direct the Human Signals lab. I hold a bachelor's degree in Computer Science from UT Austin, a master's from the MIT Media Lab and a Ph.D. in Human-Centered Computing from Georgia Tech. I am a member of DICE, WNCG, and I am currently an Associate Editor of IMWUT. I also co-lead the Life Sensing Consortium.
My research focuses on the computational perception of human signals (e.g., behavioral, emotional, physiological) while leveraging ubiquitous and wearable sensing. A core area of interest is studying systems and methods for recognizing and modeling the entire span of people's everyday activities and context. This work is at the intersection of ubiquitous computing, hci, human-centered machine learning, and signal processing. I am particularly motivated by applications in the domain of health and personalized medicine such as building health models and tools that can characterize and forecast states of health and disease from sensor data.
ethomaz at utexas dot edu
Austin, TX 78712
EE382V - Activity Sensing and Recognition
EE422C - Software Design and Implementation II
Dawei's work on the balance between audio degradation and activity recognition performance has been accepted to MobileHCI 2020. Like all other conferences in the Fall, MobileHCI will take place online.
Our paper on intra-oral sensing of motion and temperature has been accepted to ISWC 2020, which will be held online in September due to the pandemic.
Keum San's paper on eating detection with jawbone-mounted inertial sensing was featured at the 2020 IEEE Engineering in Medicine and Biology Society Conference.
We have contributed to a review paper on eating detection with wearables. Led by our collaborators at USC, it has been published at Nature Digital Medicine.
I was the workshop co-chair for PerCom 2020. Held online due to the pandemic, it was supposed to take place at UT Austin. Two papers from the group were featured in the ARDOUS workshop, focused on data annotation for Ubicomp applications.
Edison Thomaz © 2020