Parametric Representations for Nonlinear Modeling of Visual Data

Abstract

Accurate characterization of data distribution is of significant importance for vision problems. In many situations, multivariate visual data often spread into a nonlinear manifold in the high-dimensional space, which makes traditional linear modeling techniques ineffective. This paper proposes a generic nonlinear modeling scheme based on parametric data representations. We build a compact representation for the visual data using a set of parameterized basis (wavelet) functions, where the parameters are randomized to characterize the nonlinear structure of the data distribution. Meanwhile, a new progressive density approximation scheme is proposed to obtain an accurate estimate of the probability density, which imposes discrimination power on the model. Both synthetic and real image data are used to demonstrate the strength of our modeling scheme.

Cite

Text

Zhu et al. "Parametric Representations for Nonlinear Modeling of Visual Data." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2001. doi:10.1109/CVPR.2001.991011

Markdown

[Zhu et al. "Parametric Representations for Nonlinear Modeling of Visual Data." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2001.](https://mlanthology.org/cvpr/2001/zhu2001cvpr-parametric/) doi:10.1109/CVPR.2001.991011

BibTeX

@inproceedings{zhu2001cvpr-parametric,
  title     = {{Parametric Representations for Nonlinear Modeling of Visual Data}},
  author    = {Zhu, Ying and Comaniciu, Dorin and Ramesh, Visvanathan and Schwartz, Stuart C.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2001},
  pages     = {II:553-},
  doi       = {10.1109/CVPR.2001.991011},
  url       = {https://mlanthology.org/cvpr/2001/zhu2001cvpr-parametric/}
}