Improved Facial Expression Recognition via Uni-Hyperplane Classification
Abstract
Large margin learning approaches, such as support vec-tor machines (SVM), have been successfully applied to nu-merous classification tasks, especially for automatic facial expression recognition. The risk of such approaches how-ever, is their sensitivity to large margin losses due to the in-fluence from noisy training examples and outliers which is a common problem in the area of affective computing (i.e., manual coding at the frame level is tedious so coarse labels are normally assigned). In this paper, we leverage the re-laxation of the parallel-hyperplanes constraint and propose the use of modified correlation filters (MCF). The MCF is similar in spirit to SVMs and correlation filters, but with the key difference of optimizing only a single hyperplane. We demonstrate the superiority of MCF over current tech-niques on a battery of experiments. 1.
Cite
Text
Chew et al. "Improved Facial Expression Recognition via Uni-Hyperplane Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2012. doi:10.1109/CVPR.2012.6247973Markdown
[Chew et al. "Improved Facial Expression Recognition via Uni-Hyperplane Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2012.](https://mlanthology.org/cvpr/2012/chew2012cvpr-improved/) doi:10.1109/CVPR.2012.6247973BibTeX
@inproceedings{chew2012cvpr-improved,
title = {{Improved Facial Expression Recognition via Uni-Hyperplane Classification}},
author = {Chew, Sien W. and Lucey, Simon and Lucey, Patrick and Sridharan, Sridha and Conn, Jeff F.},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2012},
pages = {2554-2561},
doi = {10.1109/CVPR.2012.6247973},
url = {https://mlanthology.org/cvpr/2012/chew2012cvpr-improved/}
}