Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Linlin, Yang, Bowen, Wang, Tianhao, Li, Kangshuo, Chen, Feng
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912254075600896
author Yu, Linlin
Yang, Bowen
Wang, Tianhao
Li, Kangshuo
Chen, Feng
author_facet Yu, Linlin
Yang, Bowen
Wang, Tianhao
Li, Kangshuo
Chen, Feng
contents The fusion of raw sensor data to create a Bird's Eye View (BEV) representation is critical for autonomous vehicle planning and control. Despite the growing interest in using deep learning models for BEV semantic segmentation, anticipating segmentation errors and enhancing the explainability of these models remain underexplored. This paper introduces a comprehensive benchmark for predictive uncertainty quantification in BEV segmentation, evaluating multiple uncertainty quantification methods across three popular datasets with three representative network architectures. Our study focuses on the effectiveness of quantified uncertainty in detecting misclassified and out-of-distribution (OOD) pixels while also improving model calibration. Through empirical analysis, we uncover challenges in existing uncertainty quantification methods and demonstrate the potential of evidential deep learning techniques, which capture both aleatoric and epistemic uncertainty. To address these challenges, we propose a novel loss function, Uncertainty-Focal-Cross-Entropy (UFCE), specifically designed for highly imbalanced data, along with a simple uncertainty-scaling regularization term that improves both uncertainty quantification and model calibration for BEV segmentation.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20986
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function
Yu, Linlin
Yang, Bowen
Wang, Tianhao
Li, Kangshuo
Chen, Feng
Machine Learning
Computer Vision and Pattern Recognition
The fusion of raw sensor data to create a Bird's Eye View (BEV) representation is critical for autonomous vehicle planning and control. Despite the growing interest in using deep learning models for BEV semantic segmentation, anticipating segmentation errors and enhancing the explainability of these models remain underexplored. This paper introduces a comprehensive benchmark for predictive uncertainty quantification in BEV segmentation, evaluating multiple uncertainty quantification methods across three popular datasets with three representative network architectures. Our study focuses on the effectiveness of quantified uncertainty in detecting misclassified and out-of-distribution (OOD) pixels while also improving model calibration. Through empirical analysis, we uncover challenges in existing uncertainty quantification methods and demonstrate the potential of evidential deep learning techniques, which capture both aleatoric and epistemic uncertainty. To address these challenges, we propose a novel loss function, Uncertainty-Focal-Cross-Entropy (UFCE), specifically designed for highly imbalanced data, along with a simple uncertainty-scaling regularization term that improves both uncertainty quantification and model calibration for BEV segmentation.
title Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.20986