Chapter

Advances in Visual Computing

Volume 5876 of the series Lecture Notes in Computer Science pp 407-416

Visual Intention Detection for Wheelchair Motion

  • T. LuhandjulaAffiliated withFrench South African Technical Institute in Electronics at the Tshwane University of TechnologyMeraka Institute at the Council for Scientific and Industrial Research
  • , E. MonacelliAffiliated withLISV Laboratory, Université de Versailles St-Quentin-en-Yvelines
  • , Y. HamamAffiliated withFrench South African Technical Institute in Electronics at the Tshwane University of Technology
  • , B. J. van WykAffiliated withFrench South African Technical Institute in Electronics at the Tshwane University of Technology
  • , Q. WilliamsAffiliated withMeraka Institute at the Council for Scientific and Industrial Research

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This paper describes a visual interface that recognizes the command request of a person by inferring the intention to travel in a desired direction at a certain speed from the person’s head movements. A rotation and a vertical motion indicate the intent to change direction and speed respectively. The context for which this solution is intended is that of wheelchair bound individuals. This paper describes work in progress that provides a proof of concept tested on static images. Results show that the symmetry property of the head can be used to detect a change in its position and can therefore serve as a visual intent indicator. The solution described in this paper, focusing on the specific task of head pose estimation, intends to provide a contribution to the realisation of an enabled environment allowing people with severe disabilities and the elderly to be more independent and active in society.

Keywords

Head pose estimation intention intention detection visual interface enabled environment disabilities intention curve