CUED Publications database

Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding

Kendall, A and Badrinarayanan, V and Cipolla, R (2017) Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding. In: UNSPECIFIED.

Full text not available from this repository.

Abstract

© 2017. The copyright of this document resides with its authors. We present a deep learning framework for probabilistic pixel-wise 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. In addition, we show that modelling uncertainty improves segmentation performance by 2-3% across a number of datasets and architectures such as SegNet, FCN, Dilation Network and DenseNet.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Uncontrolled Keywords: cs.CV cs.CV cs.NE
Subjects: UNSPECIFIED
Divisions: Div F > Machine Intelligence
Depositing User: Cron Job
Date Deposited: 02 Feb 2018 20:20
Last Modified: 10 Apr 2021 23:17
DOI: