Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding

Alex Kendall, Vijay Badrinarayanan and Roberto Cipolla

Abstract

We present a deep learning framework for probabilistic pixelwise semantic segmentation, which we term Bayesian SegNet. Semantic segmentation is an important tool for visual scene understanding and a meaningful measure of uncertainty is essential for decision making. Our contribution is a practical system which is able to predict pixel-wise class labels with a measure of model uncertainty using Bayesian deep learning. We achieve this by Monte Carlo sampling with dropout at test time to generate a posterior distribution of pixel class labels.

Session

Posters

Files

PDF iconPaper (PDF)

DOI

10.5244/C.31.57
https://dx.doi.org/10.5244/C.31.57

Citation

Alex Kendall, Vijay Badrinarayanan and Roberto Cipolla. Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding. In T.K. Kim, S. Zafeiriou, G. Brostow and K. Mikolajczyk, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 57.1-57.12. BMVA Press, September 2017.

Bibtex

            @inproceedings{BMVC2017_57,
                title={Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding},
                author={Alex Kendall, Vijay Badrinarayanan and Roberto Cipolla},
                year={2017},
                month={September},
                pages={57.1-57.12},
                articleno={57},
                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.31.57},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.57}
            }