Details
| Originalsprache | Englisch |
|---|---|
| Seiten (von - bis) | 83-91 |
| Seitenumfang | 9 |
| Fachzeitschrift | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Jahrgang | 5 |
| Ausgabenummer | 2 |
| Publikationsstatus | Veröffentlicht - 17 Juni 2021 |
| Veranstaltung | 2021 24th ISPRS Congress on Imaging Today, Foreseeing Tomorrow, Commission II - Nice, Frankreich Dauer: 5 Juli 2021 → 9 Juli 2021 |
Abstract
In order to leverage and profit from unlabelled data, semi-supervised frameworks for semantic segmentation based on consistency training have been proven to be powerful tools to significantly improve the performance of purely supervised segmentation learning. However, the consensus principle behind consistency training has at least one drawback, which we identify in this paper: imbalanced label distributions within the data. To overcome the limitations of standard consistency training, we propose a novel semi-supervised framework for semantic segmentation, introducing additional losses based on prior knowledge. Specifically, we propose a lightweight architecture consisting of a shared encoder and a main decoder, which is trained in a supervised manner. An auxiliary decoder is added as additional branch in order to make use of unlabelled data based on consensus training, and we add additional constraints derived from prior information on the class distribution and on auto-encoder regularisation. Experiments performed on our concrete aggregate dataset presented in this paper demonstrate the effectiveness of the proposed approach, outperforming the segmentation results achieved by purely supervised segmentation and standard consistency training.
ASJC Scopus Sachgebiete
- Physik und Astronomie (insg.)
- Instrumentierung
- Umweltwissenschaften (insg.)
- Umweltwissenschaften (sonstige)
- Erdkunde und Planetologie (insg.)
- Erdkunde und Planetologie (sonstige)
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Jahrgang 5, Nr. 2, 17.06.2021, S. 83-91.
Publikation: Beitrag in Fachzeitschrift › Konferenzaufsatz in Fachzeitschrift › Forschung › Peer-Review
}
TY - JOUR
T1 - Semi-Supervised Segmentation of Concrete Aggregate using Consensus Regularisation and Prior Guidance
AU - Coenen, Max
AU - Schack, Tobias
AU - Beyer, Dries
AU - Heipke, Christian
AU - Haist, Michael
PY - 2021/6/17
Y1 - 2021/6/17
N2 - In order to leverage and profit from unlabelled data, semi-supervised frameworks for semantic segmentation based on consistency training have been proven to be powerful tools to significantly improve the performance of purely supervised segmentation learning. However, the consensus principle behind consistency training has at least one drawback, which we identify in this paper: imbalanced label distributions within the data. To overcome the limitations of standard consistency training, we propose a novel semi-supervised framework for semantic segmentation, introducing additional losses based on prior knowledge. Specifically, we propose a lightweight architecture consisting of a shared encoder and a main decoder, which is trained in a supervised manner. An auxiliary decoder is added as additional branch in order to make use of unlabelled data based on consensus training, and we add additional constraints derived from prior information on the class distribution and on auto-encoder regularisation. Experiments performed on our concrete aggregate dataset presented in this paper demonstrate the effectiveness of the proposed approach, outperforming the segmentation results achieved by purely supervised segmentation and standard consistency training.
AB - In order to leverage and profit from unlabelled data, semi-supervised frameworks for semantic segmentation based on consistency training have been proven to be powerful tools to significantly improve the performance of purely supervised segmentation learning. However, the consensus principle behind consistency training has at least one drawback, which we identify in this paper: imbalanced label distributions within the data. To overcome the limitations of standard consistency training, we propose a novel semi-supervised framework for semantic segmentation, introducing additional losses based on prior knowledge. Specifically, we propose a lightweight architecture consisting of a shared encoder and a main decoder, which is trained in a supervised manner. An auxiliary decoder is added as additional branch in order to make use of unlabelled data based on consensus training, and we add additional constraints derived from prior information on the class distribution and on auto-encoder regularisation. Experiments performed on our concrete aggregate dataset presented in this paper demonstrate the effectiveness of the proposed approach, outperforming the segmentation results achieved by purely supervised segmentation and standard consistency training.
KW - Auto-encoder
KW - Concrete aggregate particles
KW - Consistency training
KW - Semantic segmentation
KW - Semi-supervised learning
UR - http://www.scopus.com/inward/record.url?scp=85119673829&partnerID=8YFLogxK
U2 - 10.5194/isprs-annals-V-2-2021-83-2021
DO - 10.5194/isprs-annals-V-2-2021-83-2021
M3 - Conference article
AN - SCOPUS:85119673829
VL - 5
SP - 83
EP - 91
JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
SN - 2194-9042
IS - 2
T2 - 2021 24th ISPRS Congress on Imaging Today, Foreseeing Tomorrow, Commission II
Y2 - 5 July 2021 through 9 July 2021
ER -