ML Anthology

Motion Language of Stereo Image Sequence

Tomoya Kato, Hayato Itoh, Atsushi Imiya
CVPRW 2017 pp. 1211-1218
doi:10.1109/CVPRW.2017.160 /cvprw/2017/kato2017cvprw-motion/

Abstract

This paper proposes a method for the symbolisation of temporal changes of the environments around the autonomous robot in a work space using the scene flow field for recognition of events. The proposed algorithm is a two-phase method for the accurate computation of the scene flow field. The algorithm next symbolises the motion field as a string of words using directional statistics. Our method extracts a string of words of motion in front of the robots from images captured by a robot-mounted stereo camera system. We evaluate performance of our method using KITTI scene flow Dataset 2015. The results shows the efficiency of the method for the extraction of events in front of a robot.

CVPRW Semantic Scholar

Cite

Text

Kato et al. "Motion Language of Stereo Image Sequence." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2017. doi:10.1109/CVPRW.2017.160

Markdown

[Kato et al. "Motion Language of Stereo Image Sequence." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2017.](https://mlanthology.org/cvprw/2017/kato2017cvprw-motion/) doi:10.1109/CVPRW.2017.160

BibTeX

@inproceedings{kato2017cvprw-motion,
  title     = {{Motion Language of Stereo Image Sequence}},
  author    = {Kato, Tomoya and Itoh, Hayato and Imiya, Atsushi},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2017},
  pages     = {1211-1218},
  doi       = {10.1109/CVPRW.2017.160},
  url       = {https://mlanthology.org/cvprw/2017/kato2017cvprw-motion/}
}