Learning to Align Semantic Segmentation and 2.5d Maps for Geolocalization

Anil Armagan, Martin Hirzer, Peter M. Roth, Vincent Lepetit

CVPR 2017

doi:10.1109/CVPR.2017.488 /cvpr/2017/armagan2017cvpr-learning/

Abstract

We present an efficient method for geolocalization in urban environments starting from a coarse estimate of the location provided by a GPS and using a simple untextured 2.5D model of the surrounding buildings. Our key contribution is a novel efficient and robust method to optimize the pose: We train a Deep Network to predict the best direction to improve a pose estimate, given a semantic segmentation of the input image and a rendering of the buildings from this estimate. We then iteratively apply this CNN until converging to a good pose. This approach avoids the use of reference images of the surroundings, which are difficult to acquire and match, while 2.5D models are broadly available. We can therefore apply it to places unseen during training.

PDF CVPR Semantic Scholar

Cite

Text

Armagan et al. "Learning to Align Semantic Segmentation and 2.5d Maps for Geolocalization." Conference on Computer Vision and Pattern Recognition, 2017. doi:10.1109/CVPR.2017.488

Markdown

[Armagan et al. "Learning to Align Semantic Segmentation and 2.5d Maps for Geolocalization." Conference on Computer Vision and Pattern Recognition, 2017.](https://mlanthology.org/cvpr/2017/armagan2017cvpr-learning/) doi:10.1109/CVPR.2017.488

BibTeX

@inproceedings{armagan2017cvpr-learning,
  title     = {{Learning to Align Semantic Segmentation and 2.5d Maps for Geolocalization}},
  author    = {Armagan, Anil and Hirzer, Martin and Roth, Peter M. and Lepetit, Vincent},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2017},
  doi       = {10.1109/CVPR.2017.488},
  url       = {https://mlanthology.org/cvpr/2017/armagan2017cvpr-learning/}
}