Robust real time face tracking in mobile devices

Authors:
Elyor Kodirov;Ali Fahmi Pn;Guee-Sang Lee;Deok-Jai Choi;In Seop Na
Affiliations:
Chonnam National University, Bukgu, Gwangju, Korea;Chonnam National University, Bukgu, Gwangju, Korea;Chonnam National University, Bukgu, Gwangju, Korea;Chonnam National University, Bukgu, Gwangju, Korea;Chonnam National University, Bukgu, Gwangju, Korea
Venue:
Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication
Year:
2013

Citing 7
Cited 0

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
Contour Tracking by Stochastic Propagation of Conditional Density

ECCV '96 Proceedings of the 4th European Conference on Computer Vision-Volume I - Volume I
EigenTracking: Robust Matching and Tracking of Articulated Objects Using a View-Based Representation

ECCV '96 Proceedings of the 4th European Conference on Computer Vision-Volume I - Volume I
Kernel-Based Object Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Life Spans

IEEE Transactions on Pattern Analysis and Machine Intelligence
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking

IEEE Transactions on Signal Processing
Robust online appearance models for visual tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose an efficient algorithm for face tracking on mobile platforms. First we make an improvement over the robust mean shift tracking by introducing a new type of features, named First Class Feature Point (FCFP). And it handles normal tracking condition due to accuracy, computational cost and robustness. Secondly, in order to deal with a fast or a sudden movement of face, we combine the meanshift concept with particle filter by focusing on the localization and computational cost. In order to switch the tracking methods, that are, from improved robust meanshift tracking (IRMT) to the combination of particle tacking and meanshift (CPFMSH) and vice versa, we propose two approaches: using the difference of histograms and the velocity of the smartphone. Moreover, we compare our approach with several algorithms with several challenging video sequences. At the end, we show how experimental results of our approach can track the face very robustly, accurately and more importantly without much computational cost, with times between 6 and 20ms with respect to face size.