Siamese U-Net for Image Anomaly Detection and Segmentation with Contrastive Learning
Chia Ying Lin (National Tsing Hua University),* Shang-Hong Lai (National Tsing Hua University)
The 33rd British Machine Vision Conference
Abstract
Computing image anomaly score from the maximum of the anomaly segmentation prediction result has been widely adopted for end-to-end anomaly detection approaches. However, slight discrepancy in predicted pixel-level anomaly scores for normal and anomalous features often leads to high segmentation accuracy but unmatched poor detection performance. To overcome this problem, we propose a novel siamese-based U-Net model based on a contrastive learning framework combined with deviation-based detection finetuning strategy. The model is trained to drag normal features together while alienating the anomaly samples. Moreover, we introduce a novel channel-positional attention module (CPAM) in our U-Net decoder for refined feature upsampling. Our model reaches SOTA performance on the well-known 2D MVTecAD dataset and outperforms all other methods on the challenging dataset MVTec3D-AD by a large margin.Video
Citation
@inproceedings{Lin_2022_BMVC,
author = {Chia Ying Lin and Shang-Hong Lai},
title = {Siamese U-Net for Image Anomaly Detection and Segmentation with Contrastive Learning},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},
publisher = {{BMVA} Press},
year = {2022},
url = {https://bmvc2022.mpi-inf.mpg.de/0752.pdf}
}
Copyright © 2022
The British Machine Vision Conference is organised by