Combines pre-trained features, Bayesian regression, and moment propagation to enable real-time epistemic uncertainty for semantic segmentation on embedded systems while preserving accuracy.
Evaluating Bayesian Deep Learning Methods for Semantic Segmentation
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
Deep learning has been revolutionary for computer vision and semantic segmentation in particular, with Bayesian Deep Learning (BDL) used to obtain uncertainty maps from deep models when predicting semantic classes. This information is critical when using semantic segmentation for autonomous driving for example. Standard semantic segmentation systems have well-established evaluation metrics. However, with BDL's rising popularity in computer vision we require new metrics to evaluate whether a BDL method produces better uncertainty estimates than another method. In this work we propose three such metrics to evaluate BDL models designed specifically for the task of semantic segmentation. We modify DeepLab-v3+, one of the state-of-the-art deep neural networks, and create its Bayesian counterpart using MC dropout and Concrete dropout as inference techniques. We then compare and test these two inference techniques on the well-known Cityscapes dataset using our suggested metrics. Our results provide new benchmarks for researchers to compare and evaluate their improved uncertainty quantification in pursuit of safer semantic segmentation.
verdicts
UNVERDICTED 3representative citing papers
Monte Carlo Stochastic Depth provides a theoretically linked and empirically competitive method for uncertainty quantification in modern deep learning models such as object detectors.
Proposes MC-GLM to deliver instance-level post-hoc uncertainty for object detectors via Laplace approximation plus constant-cost Monte Carlo sampling.
citing papers explorer
-
Uncertainty in Real-Time Semantic Segmentation on Embedded Systems
Combines pre-trained features, Bayesian regression, and moment propagation to enable real-time epistemic uncertainty for semantic segmentation on embedded systems while preserving accuracy.
-
Monte Carlo Stochastic Depth for Uncertainty Estimation in Deep Learning
Monte Carlo Stochastic Depth provides a theoretically linked and empirically competitive method for uncertainty quantification in modern deep learning models such as object detectors.
-
Instance-Level Post Hoc Uncertainty Quantification in Object Detection
Proposes MC-GLM to deliver instance-level post-hoc uncertainty for object detectors via Laplace approximation plus constant-cost Monte Carlo sampling.