DMCA
Face alignment through subspace constrained mean-shifts
| Venue: | in Proc. Int. Conf. Computer Vision |
| Citations: | 93 - 10 self |
BibTeX
@INPROCEEDINGS{Saragih_facealignment,
author = {Jason M. Saragih and Simon Lucey and Jeffrey F. Cohn},
title = {Face alignment through subspace constrained mean-shifts},
booktitle = {in Proc. Int. Conf. Computer Vision},
year = {},
pages = {1034--1041}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Deformable model fitting has been actively pursued in the computer vision community for over a decade. As a result, numerous approaches have been proposed with varying degrees of success. A class of approaches that has shown substantial promise is one that makes independent predictions regarding locations of the model’s landmarks, which are combined by enforcing a prior over their joint motion. A common theme in innovations to this approach is the replacement of the distribution of probable landmark locations, obtained from each local detector, with simpler parametric forms. This simplification substitutes the true objective with a smoothed version of itself, reducing sensitivity to local minima and outlying detections. In this work, a principled optimization strategy is proposed where a nonparametric representation of the landmark distributions is maximized within a hierarchy of smoothed estimates. The resulting update equations are reminiscent of mean-shift but with a subspace constraint placed on the shape’s variability. This approach is shown to outperform other existing methods on the task of generic face fitting. 1.
Keyphrases
face alignment update equation substantial promise subspace constraint landmark distribution smoothed version common theme computer vision community joint motion local minimum generic face fitting smoothed estimate numerous approach independent prediction model landmark probable landmark location local detector deformable model fitting simpler parametric form nonparametric representation shape variability principled optimization strategy true objective
