Volume XLI-B3
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B3, 633-640, 2016
https://doi.org/10.5194/isprs-archives-XLI-B3-633-2016
© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B3, 633-640, 2016
https://doi.org/10.5194/isprs-archives-XLI-B3-633-2016
© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

  10 Jun 2016

10 Jun 2016

EFFICIENT SEMANTIC SEGMENTATION OF MAN-MADE SCENES USING FULLY-CONNECTED CONDITIONAL RANDOM FIELD

Weihao Li1 and Michael Ying Yang1,2 Weihao Li and Michael Ying Yang
  • 1TU Dresden, Computer Vision Lab Dresden, Dresden, Germany
  • 2University of Twente, ITC Faculty, EOS department, Enschede, The Netherlands

Keywords: Man-made Scene, Semantic Segmentation, Fully Connected CRFs, Mean Field Inference

Abstract. In this paper we explore semantic segmentation of man-made scenes using fully connected conditional random field (CRF). Images of man-made scenes display strong contextual dependencies in the spatial structures. Fully connected CRFs can model long-range connections within the image of man-made scenes and make use of contextual information of scene structures. The pairwise edge potentials of fully connected CRF models are defined by a linear combination of Gaussian kernels. Using filter-based mean field algorithm, the inference is very efficient. Our experimental results demonstrate that fully connected CRF performs better than previous state-of-the-art approaches on both eTRIMS dataset and LabelMeFacade dataset.