Actively Exploring Creation of Face Space(s) for Improved Face Recognition
Abstract
We propose a learning framework that actively explores cre-ation of face space(s) by selecting images that are comple-mentary to the images already represented in the face space. We also construct ensembles of classifiers learned from such actively sampled image sets, which further provides improve-ment in the recognition rates. We not only significantly re-duce the number of images required in the training set but also improve the accuracy over learning from all the images. We also show that the single face space or ensemble of face spaces, thus constructed, has a higher generalization perfor-mance across different illumination and expression condi-tions.
Cite
Text
Chawla and Bowyer. "Actively Exploring Creation of Face Space(s) for Improved Face Recognition." AAAI Conference on Artificial Intelligence, 2007.Markdown
[Chawla and Bowyer. "Actively Exploring Creation of Face Space(s) for Improved Face Recognition." AAAI Conference on Artificial Intelligence, 2007.](https://mlanthology.org/aaai/2007/chawla2007aaai-actively/)BibTeX
@inproceedings{chawla2007aaai-actively,
title = {{Actively Exploring Creation of Face Space(s) for Improved Face Recognition}},
author = {Chawla, Nitesh V. and Bowyer, Kevin W.},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2007},
pages = {809-814},
url = {https://mlanthology.org/aaai/2007/chawla2007aaai-actively/}
}