Extending Explicit Shape Regression with Mixed Feature Channels and Pose Priors
Abstract
Facial feature detection offers a wide range of applications, e.g. in facial image processing, human computer interaction, consumer electronics, and the entertainment industry. These applications impose two antagonistic key requirements: high processing speed and high detection accuracy. We address both by expanding upon the recently proposed explicit shape regression [1] to (a) allow usage and mixture of different feature channels, and (b) include head pose information to improve detection performance in non-cooperative environments. Using the publicly available âwildâ datasets LFW [10] and AFLW [11], we show that using these extensions outperforms the baseline (up to 10% gain in accuracy at 8% IOD) as well as other state-of-the-art methods.
Cite
Text
Richter et al. "Extending Explicit Shape Regression with Mixed Feature Channels and Pose Priors." IEEE/CVF Winter Conference on Applications of Computer Vision, 2014. doi:10.1109/WACV.2014.6835993Markdown
[Richter et al. "Extending Explicit Shape Regression with Mixed Feature Channels and Pose Priors." IEEE/CVF Winter Conference on Applications of Computer Vision, 2014.](https://mlanthology.org/wacv/2014/richter2014wacv-extending/) doi:10.1109/WACV.2014.6835993BibTeX
@inproceedings{richter2014wacv-extending,
title = {{Extending Explicit Shape Regression with Mixed Feature Channels and Pose Priors}},
author = {Richter, Matthias and Gao, Hua and Ekenel, Hazim Kemal},
booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
year = {2014},
pages = {1013-1019},
doi = {10.1109/WACV.2014.6835993},
url = {https://mlanthology.org/wacv/2014/richter2014wacv-extending/}
}