Robust Face Tracking with Suppressed False Positives in Smart Home Environment

Authors:
Jun-Hyeong Do;Zeungnam Bien
Affiliations:
Dept. of Medical Research, Korea Institute of Oriental Medicine, Daejeon, Korea;School of Electrical Engineering and Computer Science, KAIST, Daejeon, Korea
Venue:
ICOST '08 Proceedings of the 6th international conference on Smart Homes and Health Telematics
Year:
2008

Citing 7
Cited 0

Kernel-Based Object Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Real-Time Tracking of Multiple Persons

ICIAP '03 Proceedings of the 12th International Conference on Image Analysis and Processing
Multi-Modal Face Tracking Using Bayesian Network

AMFG '03 Proceedings of the IEEE International Workshop on Analysis and Modeling of Faces and Gestures
Robust Visual Tracking by Integrating Multiple Cues Based on Co-Inference Learning

International Journal of Computer Vision - Special Issue on Computer Vision Research at the Beckman Institute of Advanced Science and Technology
Object Tracking by the Mean-Shift of Regional Color Distribution Combined with the Particle-Filter Algorithm

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
Fast Multiple Object Tracking via a Hierarchical Particle Filter

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 - Volume 01
Real time hand tracking by combining particle filtering and mean shift

FGR' 04 Proceedings of the Sixth IEEE international conference on Automatic face and gesture recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

The face-tracking systems in use often suffer from converging to a false face-like region in a smart home environment. For this, we propose a technique to reduce errors due to false positives in the estimation scheme of face movement. In the proposed method, the face movement is estimated by using the information on face-candidate blobs obtained from the current frame as well as from the previous frame. This estimated face movement information is used in the face tracker for tracking faces in video images. Our experimental result shows a conspicuous improvement in the performance of the face tracking process in terms of success rates and with robustness against interruptions from face-like blobs.