ML Anthology

Principal Regression Analysis

Jason M. Saragih
CVPR 2011 pp. 2881-2888
doi:10.1109/CVPR.2011.5995618 /cvpr/2011/saragih2011cvpr-principal/

Abstract

A new paradigm for multivariate regression is proposed; principal regression analysis (PRA). It entails learning a low dimensional subspace over sample-specific regressors. For a given input, the model predicts a subspace thought to contain the corresponding response. Using this subspace as a prior, the search space is considerably more constrained. An efficient local optimisation strategy is proposed for learning and a practical choice for its initialisation suggested. The utility of PRA is demonstrated on the task of non-rigid face and car alignment using challenging "in the wild" datasets, where substantial performance improvements are observed over alignment with a conventional prior.

CVPR Semantic Scholar

Cite

Text

Saragih. "Principal Regression Analysis." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2011. doi:10.1109/CVPR.2011.5995618

Markdown

[Saragih. "Principal Regression Analysis." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2011.](https://mlanthology.org/cvpr/2011/saragih2011cvpr-principal/) doi:10.1109/CVPR.2011.5995618

BibTeX

@inproceedings{saragih2011cvpr-principal,
  title     = {{Principal Regression Analysis}},
  author    = {Saragih, Jason M.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2011},
  pages     = {2881-2888},
  doi       = {10.1109/CVPR.2011.5995618},
  url       = {https://mlanthology.org/cvpr/2011/saragih2011cvpr-principal/}
}