Quaternion based gesture recognition using worn inertial sensors in a motion tracking system

This work deals with the development and testing of a low latency Animation Transmission Technology (ATT) for augmented reality applications. This web-based transmission system based on AutoBahn Pyhton web server and WebSocket LibrarY (WSLAY) integrated clients has been studied to reduce packet header size and to simultaneously allow a large number of Users to interact with each other. The transmission system in the proposed system architecture allows motion sensor’s acquisition software to easily send compressed 3D quaternion-based data from transmitter to the rendering softwares at receiver side to generate real time 3D animation on an avatar. Furthermore, effects of lossy compression of quaternion data and server limitation have also been considered. Initial simulation test results with the Python server alongside WSLAY integrated clients with virtual motion sensors have been presented in this work. It has been evaluated that the network delay (ND) has a huge improvement from more than 300 ms to less than 25 ms at 100 Hz sensor sampling rate once lossy compression of 3D quaternion data is implemented. Furthermore, the effect of motion sensor sampling frequency and broadcast server limitation on maximum number of simultaneous users/sensors is also described in this work.

show abstract

Quaternion based gesture recognition using worn inertial sensors in a motion tracking system

Cited by 3 publications

References 13 publications

An Interaction Mechanism for Virtual Reality Based on Upper Limbs Motions Tracking Using Depth Cameras and Inertial Sensors

An Interaction Mechanism for Virtual Reality Based on Upper Limbs Motions Tracking Using Depth Cameras and Inertial Sensors

Gesture recognition using Markov Systems and wearable wireless inertial sensors

A low latency quaternion-based web transmission system for augmented reality applications

Contact Info

Product

Resources

About