Deformable Face Ensemble Alignment with Robust Grouped-L1 Anchors - Robotics Institute Carnegie Mellon University

Deformable Face Ensemble Alignment with Robust Grouped-L1 Anchors

Xin Cheng, Clinton Fookes, Sridha Sridharan, Jason M. Saragih, and Simon Lucey
Conference Paper, Proceedings of 10th IEEE International Conference and Workshops on Automatic Face & Gesture Recognition (FG '13), April, 2013

Abstract

Many methods exist at the moment for deformable face fitting. A drawback to nearly all these approaches is that they are (i) noisy in terms of landmark positions, and (ii) the noise is biased across frames (i.e. the misalignment is toward common directions across all frames). In this paper we propose a grouped L1-norm anchored method for simultaneously align- ing an ensemble of deformable face images stemming from the same subject, given noisy heterogeneous landmark estimates. Impressive alignment performance improvement and refine- ment is obtained using very weak initialization as “anchors”.

BibTeX

@conference{Cheng-2013-17128,
author = {Xin Cheng and Clinton Fookes and Sridha Sridharan and Jason M. Saragih and Simon Lucey},
title = {Deformable Face Ensemble Alignment with Robust Grouped-L1 Anchors},
booktitle = {Proceedings of 10th IEEE International Conference and Workshops on Automatic Face & Gesture Recognition (FG '13)},
year = {2013},
month = {April},
}