Complex Activity Recognition

From REU@MU
Revision as of 16:55, 17 January 2018 by Brylow (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Host: UbiComp Laboratory

Mentor: Dr. Iqbal Ahamed

Description: Human activity recognition has become the most widely researched topic in the field of ubiquitous computing. Ranging from phones, watches, glasses, sensors, and other wearable devices, basic human activity can be detected. But standing, walking, and sitting are just simple activities. Detecting simple activity is not sufficient to make judgments on a person's health condition, or to devise a plan for well-being. Complex activity is a more in-depth classifier of simple activities, such as cooking, driving, reading, watching TV, playing football, etc., that are not possible to identify easily. This project aims to recognize complex activity using devices equipped with various sensors. A potential system would be to use a smartphone and a digital shoe equipped with several insole sensors. Various human activities will be observed, and potential complex activity models will be created. What better way to challenge yourself to contribute in pervasive computing branches connecting healthcare, engineering, assistive technology and social sciences.

Prerequisites:

  1. Interest in computational sciences, programming, data analysis, and mathematical modeling, and
  2. Experience in programming.

References:

  1. S. Dernbach, B. Das, N. C. Krishnan, B. L. Thomas, D. J. Cook, "Simple and complex activity recognition through smart phones", Proc. 8th IE, pp. 214-221, Jun. 2012.
  2. O. Gani, A. Saha, G. Ahsan, S. Ahamed, “A Novel Framework to Recognize Complex Activity”, IEEE 41st Annual Computer Software and Applications Conference, Turin, Italy, July 2017.