DUDES: Deep Uncertainty Distillation using Ensembles for Semantic Segmentation

IF 2.1 4区 地球科学 Q3 IMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY
Steven Landgraf, Kira Wursthorn, Markus Hillemann, Markus Ulrich
{"title":"DUDES: Deep Uncertainty Distillation using Ensembles for Semantic Segmentation","authors":"Steven Landgraf, Kira Wursthorn, Markus Hillemann, Markus Ulrich","doi":"10.1007/s41064-024-00280-4","DOIUrl":null,"url":null,"abstract":"<p>The intersection of deep learning and photogrammetry unveils a critical need for balancing the power of deep neural networks with interpretability and trustworthiness, especially for safety-critical application like autonomous driving, medical imaging, or machine vision tasks with high demands on reliability. Quantifying the predictive uncertainty is a promising endeavour to open up the use of deep neural networks for such applications. Unfortunately, most current available methods are computationally expensive. In this work, we present a novel approach for efficient and reliable uncertainty estimation for semantic segmentation, which we call <b>D</b>eep <b>U</b>ncertainty <b>D</b>istillation using <b>E</b>nsembles for <b>S</b>egmentation (DUDES). DUDES applies student-teacher distillation with a Deep Ensemble to accurately approximate predictive uncertainties with a single forward pass while maintaining simplicity and adaptability. Experimentally, DUDES accurately captures predictive uncertainties without sacrificing performance on the segmentation task and indicates impressive capabilities of highlighting wrongly classified pixels and out-of-domain samples through high uncertainties on the Cityscapes and Pascal VOC 2012 dataset. With DUDES, we manage to simultaneously simplify and outperform previous work on Deep-Ensemble-based Uncertainty Distillation.</p>","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":"273 1","pages":""},"PeriodicalIF":2.1000,"publicationDate":"2024-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s41064-024-00280-4","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"IMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

The intersection of deep learning and photogrammetry unveils a critical need for balancing the power of deep neural networks with interpretability and trustworthiness, especially for safety-critical application like autonomous driving, medical imaging, or machine vision tasks with high demands on reliability. Quantifying the predictive uncertainty is a promising endeavour to open up the use of deep neural networks for such applications. Unfortunately, most current available methods are computationally expensive. In this work, we present a novel approach for efficient and reliable uncertainty estimation for semantic segmentation, which we call Deep Uncertainty Distillation using Ensembles for Segmentation (DUDES). DUDES applies student-teacher distillation with a Deep Ensemble to accurately approximate predictive uncertainties with a single forward pass while maintaining simplicity and adaptability. Experimentally, DUDES accurately captures predictive uncertainties without sacrificing performance on the segmentation task and indicates impressive capabilities of highlighting wrongly classified pixels and out-of-domain samples through high uncertainties on the Cityscapes and Pascal VOC 2012 dataset. With DUDES, we manage to simultaneously simplify and outperform previous work on Deep-Ensemble-based Uncertainty Distillation.

Abstract Image

DUDES:利用集合进行语义分割的深度不确定性蒸馏
深度学习与摄影测量学的交叉点揭示了平衡深度神经网络的强大功能与可解释性和可信度的关键需求,尤其是对于自动驾驶、医疗成像或对可靠性要求极高的机器视觉任务等安全关键型应用而言。量化预测的不确定性是将深度神经网络用于此类应用的一项大有可为的工作。遗憾的是,目前大多数可用方法的计算成本都很高。在这项工作中,我们提出了一种高效、可靠的语义分割不确定性估算新方法,我们称之为 "使用集合进行分割的深度不确定性蒸馏(DUDES)"。DUDES 利用深度集合进行学生-教师蒸馏,在保持简单性和适应性的同时,只需一次前向传递就能准确地近似预测不确定性。实验结果表明,DUDES 在不牺牲分割任务性能的情况下准确捕捉了预测不确定性,并在 Cityscapes 和 Pascal VOC 2012 数据集上通过高不确定性突出显示了错误分类的像素和域外样本,令人印象深刻。通过 DUDES,我们成功地简化了基于深度集合的不确定性蒸馏,并超越了之前的研究成果。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
8.20
自引率
2.40%
发文量
38
期刊介绍: PFG is an international scholarly journal covering the progress and application of photogrammetric methods, remote sensing technology and the interconnected field of geoinformation science. It places special editorial emphasis on the communication of new methodologies in data acquisition and new approaches to optimized processing and interpretation of all types of data which were acquired by photogrammetric methods, remote sensing, image processing and the computer-aided interpretation of such data in general. The journal hence addresses both researchers and students of these disciplines at academic institutions and universities as well as the downstream users in both the private sector and public administration. Founded in 1926 under the former name Bildmessung und Luftbildwesen, PFG is worldwide the oldest journal on photogrammetry. It is the official journal of the German Society for Photogrammetry, Remote Sensing and Geoinformation (DGPF).
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信