Detail publikace

Fast and Accurate Plane Segmentation in Depth Maps for Indoor Scenes

Originální název

Fast and Accurate Plane Segmentation in Depth Maps for Indoor Scenes

Anglický název

Fast and Accurate Plane Segmentation in Depth Maps for Indoor Scenes

Jazyk

en

Originální abstrakt

This paper deals with a scene pre-processing task - depth image segmentation. Efficiency and accuracy of several methods for depth map segmentation are explored. To meet real-time capable constraints, state-of-the-art techniques needed to be modified. Along with these modifications, new segmentation approaches are presented which aim at optimizing performance characteristics. They benefit from an assumption of human-made indoor environments by focusing on detection of planar regions. All methods were evaluated on datasets with manually annotated real environments. A comparison with alternative solutions is also presented.

Anglický abstrakt

This paper deals with a scene pre-processing task - depth image segmentation. Efficiency and accuracy of several methods for depth map segmentation are explored. To meet real-time capable constraints, state-of-the-art techniques needed to be modified. Along with these modifications, new segmentation approaches are presented which aim at optimizing performance characteristics. They benefit from an assumption of human-made indoor environments by focusing on detection of planar regions. All methods were evaluated on datasets with manually annotated real environments. A comparison with alternative solutions is also presented.

BibTex


@inproceedings{BUT91488,
  author="Rostislav {Hulík} and Vítězslav {Beran} and Michal {Španěl} and Přemysl {Kršek} and Pavel {Smrž}",
  title="Fast and Accurate Plane Segmentation in Depth Maps for Indoor Scenes",
  annote="This paper deals with a scene pre-processing task - depth image segmentation.
Efficiency and accuracy of several methods for depth map segmentation are
explored. To meet real-time capable constraints, state-of-the-art techniques
needed to be modified. Along with these modifications, new segmentation
approaches are presented which aim at optimizing performance characteristics.
They benefit from an assumption of human-made indoor environments by focusing on
detection of planar regions. All methods were evaluated on datasets with manually
annotated real environments. A comparison with alternative solutions is also
presented.",
  address="Department of Computer Graphics and Multimedia FIT BUT",
  booktitle="IEEE International Conference on Intelligent Robots and Systems",
  chapter="91488",
  doi="10.1109/IROS.2012.6385868",
  edition="NEUVEDEN",
  howpublished="electronic, physical medium",
  institution="Department of Computer Graphics and Multimedia FIT BUT",
  number="12",
  year="2012",
  month="december",
  pages="1665--1670",
  publisher="Department of Computer Graphics and Multimedia FIT BUT",
  type="conference paper"
}