Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022 |
| Herausgeber (Verlag) | Institute of Electrical and Electronics Engineers Inc. |
| Seiten | 1829-1838 |
| Seitenumfang | 10 |
| ISBN (elektronisch) | 9781665409155 |
| ISBN (Print) | 978-1-6654-0916-2 |
| Publikationsstatus | Veröffentlicht - 2022 |
| Veranstaltung | 22nd IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022 - Waikoloa, USA / Vereinigte Staaten Dauer: 3 Jan. 2022 → 8 Jan. 2022 |
Publikationsreihe
| Name | Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022 |
|---|---|
| ISSN (Print) | 2472-6737 |
| ISSN (elektronisch) | 2642-9381 |
Abstract
In industrial manufacturing processes, errors frequently occur at unpredictable times and in unknown manifestations. We tackle the problem of automatic defect detection without requiring any image samples of defective parts. Recent works model the distribution of defect-free image data, using either strong statistical priors or overly simplified data representations. In contrast, our approach handles fine-grained representations incorporating the global and local image context while flexibly estimating the density. To this end, we propose a novel fully convolutional cross-scale normalizing flow (CS-Flow) that jointly processes multiple feature maps of different scales. Using normalizing flows to assign meaningful likelihoods to input samples allows for efficient defect detection on image-level. Moreover, due to the preserved spatial arrangement the latent space of the normalizing flow is interpretable which enables to localize defective regions in the image. Our work sets a new state-of-the-art in image-level defect detection on the benchmark datasets Magnetic Tile Defects and MVTec AD showing a 100% AUROC on 4 out of 15 classes.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Maschinelles Sehen und Mustererkennung
- Informatik (insg.)
- Angewandte Informatik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022. Institute of Electrical and Electronics Engineers Inc., 2022. S. 1829-1838 (Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Fully Convolutional Cross-Scale-Flows for Image-based Defect Detection
AU - Rudolph, Marco
AU - Wehrbein, Tom
AU - Rosenhahn, Bodo
AU - Wandt, Bastian
N1 - Funding Information: Acknowledgements. This work was supported by the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor (grant no. 01DD20003), the Center for Digital Innovations (ZDIN) and the Deutsche Forschungsge-meinschaft (DFG) under Germany’s Excellence Strategy within the Cluster of Excellence PhoenixD (EXC 2122).
PY - 2022
Y1 - 2022
N2 - In industrial manufacturing processes, errors frequently occur at unpredictable times and in unknown manifestations. We tackle the problem of automatic defect detection without requiring any image samples of defective parts. Recent works model the distribution of defect-free image data, using either strong statistical priors or overly simplified data representations. In contrast, our approach handles fine-grained representations incorporating the global and local image context while flexibly estimating the density. To this end, we propose a novel fully convolutional cross-scale normalizing flow (CS-Flow) that jointly processes multiple feature maps of different scales. Using normalizing flows to assign meaningful likelihoods to input samples allows for efficient defect detection on image-level. Moreover, due to the preserved spatial arrangement the latent space of the normalizing flow is interpretable which enables to localize defective regions in the image. Our work sets a new state-of-the-art in image-level defect detection on the benchmark datasets Magnetic Tile Defects and MVTec AD showing a 100% AUROC on 4 out of 15 classes.
AB - In industrial manufacturing processes, errors frequently occur at unpredictable times and in unknown manifestations. We tackle the problem of automatic defect detection without requiring any image samples of defective parts. Recent works model the distribution of defect-free image data, using either strong statistical priors or overly simplified data representations. In contrast, our approach handles fine-grained representations incorporating the global and local image context while flexibly estimating the density. To this end, we propose a novel fully convolutional cross-scale normalizing flow (CS-Flow) that jointly processes multiple feature maps of different scales. Using normalizing flows to assign meaningful likelihoods to input samples allows for efficient defect detection on image-level. Moreover, due to the preserved spatial arrangement the latent space of the normalizing flow is interpretable which enables to localize defective regions in the image. Our work sets a new state-of-the-art in image-level defect detection on the benchmark datasets Magnetic Tile Defects and MVTec AD showing a 100% AUROC on 4 out of 15 classes.
KW - Industrial Inspection Statistical Methods
KW - Learning and Optimization
UR - http://www.scopus.com/inward/record.url?scp=85126126301&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2110.02855
DO - 10.48550/arXiv.2110.02855
M3 - Conference contribution
AN - SCOPUS:85126126301
SN - 978-1-6654-0916-2
T3 - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022
SP - 1829
EP - 1838
BT - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 22nd IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022
Y2 - 3 January 2022 through 8 January 2022
ER -