EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES
In recent years, self-supervised learning has made tremendous progress in closing the gap to supervised learning due to the rapid development of more sophisticated approaches like SimCLR, MoCo, and SwAV. However, these achievements are primarily evaluated on common benchmark datasets. In this paper,...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2022-05-01
|
Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/601/2022/isprs-archives-XLIII-B2-2022-601-2022.pdf |
_version_ | 1811232044141772800 |
---|---|
author | S. Landgraf L. Kühnlein M. Hillemann M. Hoyer S. Keller S. Keller M. Ulrich |
author_facet | S. Landgraf L. Kühnlein M. Hillemann M. Hoyer S. Keller S. Keller M. Ulrich |
author_sort | S. Landgraf |
collection | DOAJ |
description | In recent years, self-supervised learning has made tremendous progress in closing the gap to supervised learning due to the rapid development of more sophisticated approaches like SimCLR, MoCo, and SwAV. However, these achievements are primarily evaluated on common benchmark datasets. In this paper, we focus on evaluating self-supervised learning for semantic segmentation of industrial burner flames. Our goal is to build an intuition on how self-supervision performs in a scenario relevant for industrial application where training labels and the opportunities for hyperparameter tuning are limited. We demonstrate that self-supervised pre-training can constitute an alternative to the state-of-the-art approach of pre-training on ImageNet. Across all scenarios, the self-supervised approaches are less susceptible to sub-optimal learning rates and achieve higher mean accuracies than ImageNet pre-training, especially when training labels are scarce. |
first_indexed | 2024-04-12T10:55:29Z |
format | Article |
id | doaj.art-3e64224f5d0b44e9922a296ef55cc385 |
institution | Directory Open Access Journal |
issn | 1682-1750 2194-9034 |
language | English |
last_indexed | 2024-04-12T10:55:29Z |
publishDate | 2022-05-01 |
publisher | Copernicus Publications |
record_format | Article |
series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
spelling | doaj.art-3e64224f5d0b44e9922a296ef55cc3852022-12-22T03:36:06ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342022-05-01XLIII-B2-202260160710.5194/isprs-archives-XLIII-B2-2022-601-2022EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMESS. Landgraf0L. Kühnlein1M. Hillemann2M. Hoyer3S. Keller4S. Keller5M. Ulrich6Institute of Photogrammetry and Remote Sensing (IPF), Karlsruhe Institute of Technology (KIT), Germanyci-tec GmbH, GermanyInstitute of Photogrammetry and Remote Sensing (IPF), Karlsruhe Institute of Technology (KIT), Germanyci-tec GmbH, GermanyInstitute of Photogrammetry and Remote Sensing (IPF), Karlsruhe Institute of Technology (KIT), Germanyci-tec GmbH, GermanyInstitute of Photogrammetry and Remote Sensing (IPF), Karlsruhe Institute of Technology (KIT), GermanyIn recent years, self-supervised learning has made tremendous progress in closing the gap to supervised learning due to the rapid development of more sophisticated approaches like SimCLR, MoCo, and SwAV. However, these achievements are primarily evaluated on common benchmark datasets. In this paper, we focus on evaluating self-supervised learning for semantic segmentation of industrial burner flames. Our goal is to build an intuition on how self-supervision performs in a scenario relevant for industrial application where training labels and the opportunities for hyperparameter tuning are limited. We demonstrate that self-supervised pre-training can constitute an alternative to the state-of-the-art approach of pre-training on ImageNet. Across all scenarios, the self-supervised approaches are less susceptible to sub-optimal learning rates and achieve higher mean accuracies than ImageNet pre-training, especially when training labels are scarce.https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/601/2022/isprs-archives-XLIII-B2-2022-601-2022.pdf |
spellingShingle | S. Landgraf L. Kühnlein M. Hillemann M. Hoyer S. Keller S. Keller M. Ulrich EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
title | EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES |
title_full | EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES |
title_fullStr | EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES |
title_full_unstemmed | EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES |
title_short | EVALUATION OF SELF-SUPERVISED LEARNING APPROACHES FOR SEMANTIC SEGMENTATION OF INDUSTRIAL BURNER FLAMES |
title_sort | evaluation of self supervised learning approaches for semantic segmentation of industrial burner flames |
url | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/601/2022/isprs-archives-XLIII-B2-2022-601-2022.pdf |
work_keys_str_mv | AT slandgraf evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT lkuhnlein evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT mhillemann evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT mhoyer evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT skeller evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT skeller evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames AT mulrich evaluationofselfsupervisedlearningapproachesforsemanticsegmentationofindustrialburnerflames |