Head pose tracking and gesture detection using block motion vectors on mobile devices

Authors:
Renxiang Li;Cuneyt Taskiran;Mike Danielsen
Affiliations:
Motorola Labs, Schaumburg, IL;Motorola Labs, Schaumburg, IL;Motorola Labs, Schaumburg, IL
Venue:
Mobility '07 Proceedings of the 4th international conference on mobile technology, applications, and systems and the 1st international symposium on Computer human interaction in mobile technology
Year:
2007

Citing 5
Cited 1

Fast, Reliable Head Tracking under Varying Illumination: An Approach Based on Registration of Texture-Mapped 3D Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Comparative Study of Coarse Head Pose Estimation

MOTION '02 Proceedings of the Workshop on Motion and Video Computing
A Stabilized Adaptive Appearance Changes Model for 3D Head Tracking

RATFG-RTS '01 Proceedings of the IEEE ICCV Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems (RATFG-RTS'01)
A real-time head nod and shake detector

Proceedings of the 2001 workshop on Perceptive user interfaces
The Smart Phone: A Ubiquitous Input Device

IEEE Pervasive Computing

Design considerations of expressive bidirectional telepresence robots

CHI '11 Extended Abstracts on Human Factors in Computing Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a novel and practical algorithm for head pose tracking and head gesture detection applicable to avatar applications and Human Computer Interaction (HCI) on mobile devices. The algorithm takes advantage of block motion vectors estimated for real-time video encoding on the device. After spatial and temporal smoothing, block motion vectors are mapped into a video pointer signal, which is further mapped into head pose signals for avatar animation control. In contrast to conventional tracking algorithms, our algorithm processes block motion vectors rather than pixel data, leading to drastically reduced computational requirement. This makes it a practical solution for head tracking in real time on high end mobile devices. A simple and reliable way of detecting head nod and shake gestures is also presented, using a deterministic finite state machine.