Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving
Deep neural networks (DNNs) have proven their capabilities in the past years and play a significant role in environment perception for the challenging application of automated driving. They are employed for tasks such as detection, semantic segmentation, and sensor fusion. Despite tremendous researc...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10128983/ |
_version_ | 1797808509965828096 |
---|---|
author | Manuel Schwonberg Joshua Niemeijer Jan-Aike Termohlen Jorg P. Schafer Nico M. Schmidt Hanno Gottschalk Tim Fingscheidt |
author_facet | Manuel Schwonberg Joshua Niemeijer Jan-Aike Termohlen Jorg P. Schafer Nico M. Schmidt Hanno Gottschalk Tim Fingscheidt |
author_sort | Manuel Schwonberg |
collection | DOAJ |
description | Deep neural networks (DNNs) have proven their capabilities in the past years and play a significant role in environment perception for the challenging application of automated driving. They are employed for tasks such as detection, semantic segmentation, and sensor fusion. Despite tremendous research efforts, several issues still need to be addressed that limit the applicability of DNNs in automated driving. The bad generalization of DNNs to unseen domains is a major problem on the way to a safe, large-scale application, because manual annotation of new domains is costly, particularly for semantic segmentation. For this reason, methods are required to adapt DNNs to new domains without labeling effort. This task is termed unsupervised domain adaptation (UDA). While several different domain shifts challenge DNNs, the shift between synthetic and real data is of particular importance for automated driving, as it allows the use of simulation environments for DNN training. We present an overview of the current state of the art in this research field. We categorize and explain the different approaches for UDA. The number of considered publications is larger than any other survey on this topic. We also go far beyond the description of the UDA state-of-the-art, as we present a quantitative comparison of approaches and point out the latest trends in this field. We conduct a critical analysis of the state-of-the-art and highlight promising future research directions. With this survey, we aim to facilitate UDA research further and encourage scientists to exploit novel research directions. |
first_indexed | 2024-03-13T06:38:36Z |
format | Article |
id | doaj.art-eff12d6af3374522b5feff09b79894c1 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-13T06:38:36Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-eff12d6af3374522b5feff09b79894c12023-06-08T23:01:01ZengIEEEIEEE Access2169-35362023-01-0111542965433610.1109/ACCESS.2023.327778510128983Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated DrivingManuel Schwonberg0https://orcid.org/0009-0002-0167-6902Joshua Niemeijer1Jan-Aike Termohlen2https://orcid.org/0009-0001-0960-6848Jorg P. Schafer3Nico M. Schmidt4Hanno Gottschalk5https://orcid.org/0000-0003-2167-2028Tim Fingscheidt6https://orcid.org/0000-0002-8895-5041CARIAD SE, Wolfsburg, GermanyDeutsches Zentrum für Luft- und Raumfahrt (DLR) e.V, Braunschweig, GermanyInstitute for Communications Technology, Technische Universität Braunschweig, Braunschweig, GermanyDeutsches Zentrum für Luft- und Raumfahrt (DLR) e.V, Braunschweig, GermanyCARIAD SE, Wolfsburg, GermanyInstitute of Mathematics, Technical University Berlin, Berlin, GermanyInstitute for Communications Technology, Technische Universität Braunschweig, Braunschweig, GermanyDeep neural networks (DNNs) have proven their capabilities in the past years and play a significant role in environment perception for the challenging application of automated driving. They are employed for tasks such as detection, semantic segmentation, and sensor fusion. Despite tremendous research efforts, several issues still need to be addressed that limit the applicability of DNNs in automated driving. The bad generalization of DNNs to unseen domains is a major problem on the way to a safe, large-scale application, because manual annotation of new domains is costly, particularly for semantic segmentation. For this reason, methods are required to adapt DNNs to new domains without labeling effort. This task is termed unsupervised domain adaptation (UDA). While several different domain shifts challenge DNNs, the shift between synthetic and real data is of particular importance for automated driving, as it allows the use of simulation environments for DNN training. We present an overview of the current state of the art in this research field. We categorize and explain the different approaches for UDA. The number of considered publications is larger than any other survey on this topic. We also go far beyond the description of the UDA state-of-the-art, as we present a quantitative comparison of approaches and point out the latest trends in this field. We conduct a critical analysis of the state-of-the-art and highlight promising future research directions. With this survey, we aim to facilitate UDA research further and encourage scientists to exploit novel research directions.https://ieeexplore.ieee.org/document/10128983/Computer visiondeep neural networksunsupervised domain adaptationsemantic segmentationautomated driving |
spellingShingle | Manuel Schwonberg Joshua Niemeijer Jan-Aike Termohlen Jorg P. Schafer Nico M. Schmidt Hanno Gottschalk Tim Fingscheidt Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving IEEE Access Computer vision deep neural networks unsupervised domain adaptation semantic segmentation automated driving |
title | Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving |
title_full | Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving |
title_fullStr | Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving |
title_full_unstemmed | Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving |
title_short | Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving |
title_sort | survey on unsupervised domain adaptation for semantic segmentation for visual perception in automated driving |
topic | Computer vision deep neural networks unsupervised domain adaptation semantic segmentation automated driving |
url | https://ieeexplore.ieee.org/document/10128983/ |
work_keys_str_mv | AT manuelschwonberg surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT joshuaniemeijer surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT janaiketermohlen surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT jorgpschafer surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT nicomschmidt surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT hannogottschalk surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving AT timfingscheidt surveyonunsuperviseddomainadaptationforsemanticsegmentationforvisualperceptioninautomateddriving |