ML Anthology

Background Segmentation with Feedback: The Pixel-Based Adaptive Segmenter

Martin Hofmann, Philipp Tiefenbacher, Gerhard Rigoll
CVPRW 2012 pp. 38-43
doi:10.1109/CVPRW.2012.6238925 /cvprw/2012/hofmann2012cvprw-background/

Abstract

In this paper we present a novel method for foreground segmentation. Our proposed approach follows a non-parametric background modeling paradigm, thus the background is modeled by a history of recently observed pixel values. The foreground decision depends on a decision threshold. The background update is based on a learning parameter. We extend both of these parameters to dynamic per-pixel state variables and introduce dynamic controllers for each of them. Furthermore, both controllers are steered by an estimate of the background dynamics. In our experiments, the proposed Pixel-Based Adaptive Segmenter (PBAS) outperforms most state-of-the-art methods.

PDF CVPRW Semantic Scholar

Cite

Text

Hofmann et al. "Background Segmentation with Feedback: The Pixel-Based Adaptive Segmenter." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2012. doi:10.1109/CVPRW.2012.6238925

Markdown

[Hofmann et al. "Background Segmentation with Feedback: The Pixel-Based Adaptive Segmenter." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2012.](https://mlanthology.org/cvprw/2012/hofmann2012cvprw-background/) doi:10.1109/CVPRW.2012.6238925

BibTeX

@inproceedings{hofmann2012cvprw-background,
  title     = {{Background Segmentation with Feedback: The Pixel-Based Adaptive Segmenter}},
  author    = {Hofmann, Martin and Tiefenbacher, Philipp and Rigoll, Gerhard},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2012},
  pages     = {38-43},
  doi       = {10.1109/CVPRW.2012.6238925},
  url       = {https://mlanthology.org/cvprw/2012/hofmann2012cvprw-background/}
}