Application of Feature Tracking in a Vision Based Human Machine Interface for XBOX

被引：3

作者：

Oskoei, Mohammadreza Asghari ^{[1
]}

Hu, Huosheng ^{[1
]}

机构：

[1] Univ Essex, Sch Comp Sci & Elect Engn, Colchester CO4 3SQ, Essex, England

来源：

2009 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO 2009), VOLS 1-4 | 2009年

关键词：

Computer Vision; Optical Flow; Motion Tracking; Human Machine Interface; Xbox;

D O I：

10.1109/ROBIO.2009.5420435

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a vision based human machine interface (HMI) for the Xbox. It applies feature tracking algorithms to recognize user's head gestures and translates them into commands for the game. The pyramidal implementation of Lucas Kanade feature tracking is used to trace the optical flows in a sequence of frames. The experimental results show the feasibility of the proposed vision based interface, which has reasonable performance comparing with the standard game pad.

引用

页码：1738 / 1743

页数：6