Few-Shot Defect Segmentation Leveraging Abundant Defect-Free Training Samples Through Normal Background Regularization And Crop-And-Paste Operation

Few-Shot Defect Segmentation Leveraging Abundant Defect-Free Training Samples Through Normal Background Regularization And Crop-And-Paste Operation
Title:
Few-Shot Defect Segmentation Leveraging Abundant Defect-Free Training Samples Through Normal Background Regularization And Crop-And-Paste Operation
Other Titles:
2021 IEEE International Conference on Multimedia and Expo (ICME)
Keywords:
Publication Date:
09 June 2021
Citation:
Lin, D., Cao, Y., Zhu, W., & Li, Y. (2021). Few-Shot Defect Segmentation Leveraging Abundant Defect-Free Training Samples Through Normal Background Regularization And Crop-And-Paste Operation. 2021 IEEE International Conference on Multimedia and Expo (ICME). doi:10.1109/icme51207.2021.9428468
Abstract:
In industrial quality assessment, it is challenging to conduct automated and accurate defect segmentation under the condition that abundant defect-free images but very limited anomalous images are available. This paper tackles the challenging few-shot defect segmentation task under such condition. We propose two regularization techniques via incorporating abundant defect-free images into the training of an encoder-decoder segmentation network. We first propose a Normal Background Regularization (NBR) loss which is jointly minimized with the segmentation loss, enhancing the encoder network to produce discriminative representations for normal regions. Secondly, we crop/paste defective regions to the randomly selected normal images for data augmentation and propose a weighted binary cross-entropy loss to enhance the training by emphasizing more realistic crop-and-pasted augmented images based on feature-level similarity comparison. Extensive experiments on MVTec AD and MTSD datasets demonstrate the superiority of the proposed method over the competing methods under few-shot settings.
License type:
Publisher Copyright
Funding Info:
There was no specific funding for the research done
Description:
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
1945-788X
1945-7871
ISBN:
978-1-6654-1152-3
978-1-6654-3864-3
Files uploaded:

File Size Format Action
icme2021template.pdf 3.07 MB PDF Request a copy