Efficient 3D Tracking in Urban Environments with Semantic Segmentation

Martin Hirzer, Peter Roth and Vincent Lepetit

Abstract

In this paper, we present a new 3D tracking approach for self-localization in urban environments. In particular, we build on existing tracking approaches (i.e., visual odometry tracking and SLAM), additionally using the information provided by 2.5D maps of the environment. Since this combination is not straightforward, we adopt ideas from semantic segmentation to find a better alignment between the pose estimated by the tracker and the 2.5D model. Specifically, we show that introducing edges as semantic classes is highly beneficial for our task. In this way, we can reduce tracker inaccuracies and prevent drifting, thus increasing the tracker’s stability.

Session

Posters

Files

PDF iconPaper (PDF)
PDF iconSupplementary (PDF)

DOI

10.5244/C.31.143
https://dx.doi.org/10.5244/C.31.143

Citation

Martin Hirzer, Peter Roth and Vincent Lepetit. Efficient 3D Tracking in Urban Environments with Semantic Segmentation. In T.K. Kim, S. Zafeiriou, G. Brostow and K. Mikolajczyk, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 143.1-143.12. BMVA Press, September 2017.

Bibtex

            @inproceedings{BMVC2017_143,
                title={Efficient 3D Tracking in Urban Environments with Semantic Segmentation},
                author={Martin Hirzer, Peter Roth and Vincent Lepetit},
                year={2017},
                month={September},
                pages={143.1-143.12},
                articleno={143},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.143},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.143}
            }