Demo abstract: Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data

Shahriar Nirjon, Chris Greenwood, Carlos Torres, Stefanie Zhou, John A. Stankovic, Hee Jung Yoon, Ho Kyeong Ra, Can Basaran, Taejoon Park, Sang H. Son

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Kintense is a robust, accurate, real-time, and evolving system for detecting aggressive actions such as hitting, kicking, pushing, and throwing from streaming 3D skeleton joint coordinates obtained from Kinect sensors. Kintense uses a combination of: (1) an array of supervised learners to recognize a predefined set of aggressive actions, (2) an unsupervised learner to discover new aggressive actions or refine existing actions, and (3) human feedback to reduce false alarms and to label potential aggressive actions. This abstract provides an overview of the design and implementation of Kintense and provides empirical evidence that Kintense is 11%-16% more accurate when compared to standard techniques such as dynamic time warping (DTW) and posture based gesture recognizers.

Original languageEnglish
Title of host publicationSenSys 2013 - Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems
PublisherAssociation for Computing Machinery
ISBN (Print)9781450320276
DOIs
StatePublished - 2013
Event11th ACM Conference on Embedded Networked Sensor Systems, SenSys 2013 - Rome, Italy
Duration: 11 Nov 201315 Nov 2013

Publication series

NameSenSys 2013 - Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems

Conference

Conference11th ACM Conference on Embedded Networked Sensor Systems, SenSys 2013
Country/TerritoryItaly
CityRome
Period11/11/1315/11/13

Keywords

  • Aggressive Actions
  • Kinect
  • Skeletal Tracking

Fingerprint

Dive into the research topics of 'Demo abstract: Kintense: A robust, accurate, real-time and evolving system for detecting aggressive actions from streaming 3D skeleton data'. Together they form a unique fingerprint.

Cite this