ML Anthology

Generative Modeling of Spatio-Temporal Traffic Sign Trajectories

Karla Brkic, Sinisa Segvic, Zoran Kalafatic, Ivan Sikiric, Axel Pinz
CVPRW 2010 pp. 25-31
doi:10.1109/CVPRW.2010.5543888 /cvprw/2010/brkic2010cvprw-generative/

Abstract

We consider the task of automatic detection and recognition of traffic signs in video. We show that successful off-the-shelf detection (Viola-Jones) and classification (SVM) systems yield unsatisfactory results. Our main concern are high false positive detection rates which occur due to sparseness of the traffic signs in videos. We address the problem by enforcing spatio-temporal consistency of the detections corresponding to a distinct sign in video. We also propose a generative model of the traffic sign motion in the image plane, which is obtained by clustering the trajectories filtered by an appropriate procedure. The contextual information recovered by the proposed model will be employed in our future research on recognizing traffic signs in video.

PDF CVPRW Semantic Scholar

Cite

Text

Brkic et al. "Generative Modeling of Spatio-Temporal Traffic Sign Trajectories." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2010. doi:10.1109/CVPRW.2010.5543888

Markdown

[Brkic et al. "Generative Modeling of Spatio-Temporal Traffic Sign Trajectories." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2010.](https://mlanthology.org/cvprw/2010/brkic2010cvprw-generative/) doi:10.1109/CVPRW.2010.5543888

BibTeX

@inproceedings{brkic2010cvprw-generative,
  title     = {{Generative Modeling of Spatio-Temporal Traffic Sign Trajectories}},
  author    = {Brkic, Karla and Segvic, Sinisa and Kalafatic, Zoran and Sikiric, Ivan and Pinz, Axel},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2010},
  pages     = {25-31},
  doi       = {10.1109/CVPRW.2010.5543888},
  url       = {https://mlanthology.org/cvprw/2010/brkic2010cvprw-generative/}
}