Skip to main content

Research Repository

Advanced Search

L 2, 1-based regression and prediction accumulation across views for robust facial landmark detection

Martinez, Brais; Valstar, Michel F.

Authors

Brais Martinez



Abstract

We propose a new methodology for facial landmark detection. Similar to other state-of-the-art methods, we rely on the use of cascaded regression to perform inference, and we use a feature representation that results from concatenating 66 HOG descriptors, one per landmark. However, we propose a novel regression method that substitutes the commonly used Least Squares regressor. This new method makes use of the L2,1 norm, and it is designed to increase the robust- ness of the regressor to poor initialisations (e.g., due to large out of plane head poses) or partial occlusions. Furthermore, we propose to use multiple initialisations, consisting of both spatial translation and 4 head poses corresponding to different pan rotations. These estimates are aggregated into a single prediction in a robust manner. Both strategies are designed to improve the convergence behaviour of the algorithm, so that it can cope with the challenges of in-the- wild data. We further detail some important experimental details, and show extensive performance comparisons highlighting the performance improvement attained by the method proposed here.

Journal Article Type Article
Publication Date 2016-03
Journal Image and Vision Computing
Print ISSN 0262-8856
Electronic ISSN 0262-8856
Publisher Elsevier
Peer Reviewed Peer Reviewed
APA6 Citation Martinez, B., & Valstar, M. F. (2016). L 2, 1-based regression and prediction accumulation across views for robust facial landmark detection. Image and Vision Computing, doi:10.1016/j.imavis.2015.09.003
DOI https://doi.org/10.1016/j.imavis.2015.09.003
Keywords Facial landmark detection, Regression, 300 W challenge
Publisher URL http://www.sciencedirect.com/science/article/pii/S0262885615001092
Copyright Statement Copyright information regarding this work can be found at the following address: http://creativecommons.org/licenses/by-nc-nd/4.0

Files

2015IVC_L21.pdf (2.9 Mb)
PDF

Copyright Statement
Copyright information regarding this work can be found at the following address: http://creativecommons.org/licenses/by-nc-nd/4.0





You might also like



Downloadable Citations

;