Hand gesture recognition for TV remote control using tree-based ensemble and LCS classifiers

Helman Stern, Darya Frolova, Sigal Berman

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

6 Scopus citations

Abstract

The aim of this research to develop a real-time, robust, vision based hand-gesture control system capable of operating in different living room environments including low light conditions, and using inexpensive hardware components. The system is designed for remote control of a TV set-top box using a preset gesture vocabulary. A camera is located above a TV screen, across from the user. Gestures may be performed from anywhere within the sensors field of view, without the requirement that the operator obtain visual or auditory feedback. The system is capable of using either a near-IR or a visible color webcam. Hand tracking is based on a modified CAMShift algorithm with face detection and blocking. Gesture classification is based on a limited number of gesture motion trajectory features using a tree-based classifier ensemble, and a modified Longest Common Subsequence (LCS) method. Preliminary testing and evaluation of the systems real time operation is reported.

Original languageEnglish
Title of host publicationProceedings of the 2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010
Pages687-693
Number of pages7
Volume2
StatePublished - 1 Dec 2010
Event2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010 - Las Vegas, NV, United States
Duration: 12 Jul 201015 Jul 2010

Conference

Conference2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010
Country/TerritoryUnited States
CityLas Vegas, NV
Period12/07/1015/07/10

Keywords

  • Gesture recognition
  • Hand tracking
  • LCS
  • TV control
  • Tree-based ensemble classifier

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Hand gesture recognition for TV remote control using tree-based ensemble and LCS classifiers'. Together they form a unique fingerprint.

Cite this