Sampling Representative Examples for Dimensionality Reduction and Recognition - Bootstrap Bumping LDA
Abstract
We present a novel method for dimensionality reduction and recognition based on Linear Discriminant Analysis (LDA), which specifically deals with the Small Sample Size (SSS) problem in Computer Vision applications. Unlike the traditional methods, which impose specific assumptions to address the SSS problem, our approach introduces a variant of bootstrap bumping technique, which is a general framework in statistics for model search and inference. An intermediate linear representation is first hypothesized from each bootstrap sample. Then LDA is performed in the reduced subspace. Lastly, the final model is selected among all hypotheses for the best classification. Experiments on synthetic and real datasets demonstrate the advantages of our Bootstrap Bumping LDA (BB-LDA) approach over the traditional LDA based methods.
Cite
Text
Gao and Davis. "Sampling Representative Examples for Dimensionality Reduction and Recognition - Bootstrap Bumping LDA." European Conference on Computer Vision, 2006. doi:10.1007/11744078_22Markdown
[Gao and Davis. "Sampling Representative Examples for Dimensionality Reduction and Recognition - Bootstrap Bumping LDA." European Conference on Computer Vision, 2006.](https://mlanthology.org/eccv/2006/gao2006eccv-sampling/) doi:10.1007/11744078_22BibTeX
@inproceedings{gao2006eccv-sampling,
title = {{Sampling Representative Examples for Dimensionality Reduction and Recognition - Bootstrap Bumping LDA}},
author = {Gao, Hui and Davis, James W.},
booktitle = {European Conference on Computer Vision},
year = {2006},
pages = {275-287},
doi = {10.1007/11744078_22},
url = {https://mlanthology.org/eccv/2006/gao2006eccv-sampling/}
}