Experiential systems, health and well-being

We are strongly motivated by the applications of our algorithmic innovations in the emerging area of interactive experiential systems: as diverse as real-time feedback for stroke and Parkinson’s rehab, to media-systems for general well-being, to next generation of interactive performative techniques. We collaborate widely with researchers in bio-mechanics, exercise science, physical therapy, and media-arts. In many of these application, our core approach is to model underlying dynamical processes relevant to the human activity of interest, and extract interpretable data streams which can be used to drive real-time feedback via sonification, haptification, and other diverse forms of sensorial feedback. Some of our sample work in this area is given below.

We how how (elastic) temporal/timing variability in human activity can be formally handled in machine learning pipelines with functional analysis methods. Out methods push performance on accelerometer-based activity recognition by a significant margin.
  • Hongjun Choi , Qiao Wang, Meynard Toledo, Pavan Turaga, Matthew Buman, Anuj Srivastava, “Temporal Alignment Improves Feature Quality: an Experiment on Activity Recognition with Accelerometer Data”, 4th International Workshop on Differential Geometry in Computer Vision and Machine Learning (DiffCVML) in conjunction with IEEE CVPR 2018.

Home-based stroke rehab system, from which data was used in development of new dynamical modeling and quality analysis techniques.  (Venkataraman et al JBHI 2016)
  • Vinay Venkataraman, Pavan Turaga, Michael Baran, Nicole Lehrer, Tingfang Du, Long Cheng, Thanassis Rikakis, and Steven L. Wolf, “Component-Level Tuning of Kinematic Features from Composite Therapist Impressions of Movement Quality”, at IEEE Journal on Biomedical and Health Informatics (J-BHI) 2016. [pdf]
  • Michael Baran, Nicole Lehrer, Margaret Duff, Vinay Venkataraman, Pavan Turaga, Todd Ingalls, Zev Rymer, Steven L. Wolf, and Thanassis Rikakis, “Interdisciplinary Concepts for Design and Implementation of Mixed Reality Interactive Neurorehabilitation Systems for Stroke”, at the American Physical Therapy Association’s Physical Therapy Journal (APTA-PTJ) 2015. [pdf]
  • V. Venkataraman, P. Turaga, N. Lehrer, M. Baran, T. Rikakis, S. L. Wolf, “Decision Support for Stroke Rehabilitation Therapy via Describable Attribute-based Decision Trees”, in International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), August 2014. [pdf]

We have developed real-time feedback systems as well as analysis methods rooted in geometry to study the quality of day-to-day movements for preventive interventions. (Wang et al CHI 2014, Som et al, DiffCVML 2016)
  • Q. Wang, P. Turaga, G. Coleman, T. Ingalls, “SomaTech: An Exploratory Interface for Altering Movement Habits”, in ACM CHI Conference on Human Factors in Computing Systems Extended Abstracts, April 2014. [pdf]

Youtube Link

  • A. Som, R. Anirudh, Q. Wang, P. Turaga, “Riemannian Geometric Approaches for Measuring Movement Quality”, 2nd International workshop on Differential Geometry in Computer Vision and Machine Learning (DiffCVML), held in conjunction with CVPR 2016.


We collaborate with media-artists to explore computational problems that arise in the study of next-generation interactive embodied performance (Iyengar et al MOCO 2016, Krzyzaniak et al MOCO 2015).
  • Varsha Iyengar, Grisha Coleman, David Tinapple, and Pavan Turaga. 2016. Motion, Captured: an Open Repository for Comparative Movement Studies. In Proceedings of the 3rd International Symposium on Movement and Computing (MOCO ’16).
  • Michael Krzyzaniak, Rushil Anirudh, Vinay Venkataraman, Pavan Turaga and Xin Wei Sha, “Towards Realtime Measurement of Connectedness in Human Movement”,  in 2nd International Workshop on Movement and Computing (MOCO), August 2015.