TY - GEN

T1 - Same Same But DifferNet

T2 - IEEE Winter Conference on Applications of Computer Vision, WACV 2021

AU - Rudolph, Marco

AU - Wandt, Bastian

AU - Rosenhahn, Bodo

N1 - Funding Information: This work was funded by the Deutsche Forschungsgemein-schaft (DFG, German Research Foundation) under Germany’s Excellence Strategy within the Cluster of Excellence PhoenixD (EXC 2122).

PY - 2021

Y1 - 2021

N2 - The detection of manufacturing errors is crucial in fabrication processes to ensure product quality and safety standards. Since many defects occur very rarely and their characteristics are mostly unknown a priori, their detection is still an open research question. To this end, we propose DifferNet: It leverages the descriptiveness of features extracted by convolutional neural networks to estimate their density using normalizing flows. Normalizing flows are well-suited to deal with low dimensional data distributions. However, they struggle with the high dimensionality of images. Therefore, we employ a multi-scale feature extractor which enables the normalizing flow to assign meaningful likelihoods to the images. Based on these likelihoods we develop a scoring function that indicates defects. Moreover, propagating the score back to the image enables pixel-wise localization. To achieve a high robustness and performance we exploit multiple transformations in training and evaluation. In contrast to most other methods, ours does not require a large number of training samples and performs well with as low as 16 images. We demonstrate the superior performance over existing approaches on the challenging and newly proposed MVTec AD and Magnetic Tile Defects datasets.

AB - The detection of manufacturing errors is crucial in fabrication processes to ensure product quality and safety standards. Since many defects occur very rarely and their characteristics are mostly unknown a priori, their detection is still an open research question. To this end, we propose DifferNet: It leverages the descriptiveness of features extracted by convolutional neural networks to estimate their density using normalizing flows. Normalizing flows are well-suited to deal with low dimensional data distributions. However, they struggle with the high dimensionality of images. Therefore, we employ a multi-scale feature extractor which enables the normalizing flow to assign meaningful likelihoods to the images. Based on these likelihoods we develop a scoring function that indicates defects. Moreover, propagating the score back to the image enables pixel-wise localization. To achieve a high robustness and performance we exploit multiple transformations in training and evaluation. In contrast to most other methods, ours does not require a large number of training samples and performs well with as low as 16 images. We demonstrate the superior performance over existing approaches on the challenging and newly proposed MVTec AD and Magnetic Tile Defects datasets.

KW - cs.CV

KW - cs.LG

KW - eess.IV

UR - http://www.scopus.com/inward/record.url?scp=85116166966&partnerID=8YFLogxK

U2 - 10.1109/WACV48630.2021.00195

DO - 10.1109/WACV48630.2021.00195

M3 - Conference contribution

SN - 978-1-6654-0477-8

T3 - Proceedings - 2015 IEEE Winter Conference on Applications of Computer Vision Workshops, WACVW 2021

SP - 1906

EP - 1915

BT - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021

Y2 - 5 January 2021 through 9 January 2021

ER -