Fusing Geometry and Appearance for Road Segmentation

Abstract

We propose a novel method for fusing geometric and appearance cues for road surface segmentation. Modeling colour cues using Gaussian mixtures allows the fusion to be performed optimally within a Bayesian framework, avoiding ad hoc weights. Adaptation to different scene conditions is accomplished through nearest-neighbour appearance model selection over a dictionary of mixture models learned from training data, and the thorny problem of selecting the number of components in each mixture is solved through a novel cross-validation approach. Quantitative evaluation reveals that the proposed fusion method significantly improves segmentation accuracy relative to a method that uses geometric cues alone.

Cite

Text

Cheng et al. "Fusing Geometry and Appearance for Road Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2017. doi:10.1109/ICCVW.2017.28

Markdown

[Cheng et al. "Fusing Geometry and Appearance for Road Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2017.](https://mlanthology.org/iccvw/2017/cheng2017iccvw-fusing/) doi:10.1109/ICCVW.2017.28

BibTeX

@inproceedings{cheng2017iccvw-fusing,
  title     = {{Fusing Geometry and Appearance for Road Segmentation}},
  author    = {Cheng, Gong and Qian, Yiming and Elder, James H.},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2017},
  pages     = {166-173},
  doi       = {10.1109/ICCVW.2017.28},
  url       = {https://mlanthology.org/iccvw/2017/cheng2017iccvw-fusing/}
}