elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Best Practices in Active Learning for Semantic Segmentation

Niemeijer, Joshua and Mittal, Sudhanshu and Schäfer, Jörg P. and Brox, Thomas (2023) Best Practices in Active Learning for Semantic Segmentation. In: 45th Annual Conference of the German Association for Pattern Recognition. German Conference on Pattern Recognition (GCPR), 2023-09-19 - 2023-09-22, Heidelberg. doi: 10.1007/978-3-031-54605-1_28. ISBN 978-303154604-4. ISSN 03029743.

[img] PDF
2MB

Abstract

Active learning is particularly of interest for semantic segmentation, where annotations are costly. Previous academic studies focused on datasets that are already very diverse and where the model is trained in a supervised manner with a large annotation budget. In contrast, data collected in many driving scenarios is highly redundant, and most medical applications are subject to very constrained annotation budgets. This work investigates the various types of existing active learning methods for semantic segmentation under diverse conditions across three dimensions - data distribution w.r.t. different redundancy levels, integration of semi-supervised learning, and different labeling budgets. We find that these three underlying factors are decisive for the selection of the best active learning approach. As an outcome of our study, we provide a comprehensive usage guide to obtain the best performance for each case. It is the first systematic study that investigates these dimensions covering a wide range of settings including more than 3K model training runs. In this work, we also propose an exemplary evaluation task for driving scenarios, where data has high redundancy, to showcase the practical implications of our research findings.

Item URL in elib:https://elib.dlr.de/198546/
Document Type:Conference or Workshop Item (Speech)
Title:Best Practices in Active Learning for Semantic Segmentation
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Niemeijer, JoshuaUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Mittal, SudhanshuUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Schäfer, Jörg P.UNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Brox, ThomasUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Date:September 2023
Journal or Publication Title:45th Annual Conference of the German Association for Pattern Recognition
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:No
DOI:10.1007/978-3-031-54605-1_28
ISSN:03029743
ISBN:978-303154604-4
Status:Published
Keywords:Active Learning, Semantic Segmentation
Event Title:German Conference on Pattern Recognition (GCPR)
Event Location:Heidelberg
Event Type:international Conference
Event Start Date:19 September 2023
Event End Date:22 September 2023
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Road Transport
DLR - Research area:Transport
DLR - Program:V ST Straßenverkehr
DLR - Research theme (Project):V - KoKoVI - Koordinierter kooperativer Verkehr mit verteilter, lernender Intelligenz
Location: Berlin-Adlershof , Braunschweig
Institutes and Institutions:Institute of Transportation Systems
Institute of Transportation Systems > Cooperative Systems, BS
Institute of Transportation Systems > Cooperative Systems, BA
Deposited By: Niemeijer, Joshua
Deposited On:05 Dec 2023 14:30
Last Modified:13 May 2024 17:45

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.