notesum.ai

Published at November 27

Utilizing the Mean Teacher with Supcontrast Loss for Wafer Pattern Recognition

cs.CV

Released Date: November 27, 2024

Authors: Qiyu Wei¹, Xun Xu², Zeng Zeng¹, Xulei Yang²

Aff.: ¹School of Microelectronics, Shanghai University, Shanghai, China; ²Institute for Infocomm Research, A*Star, Singapore, Singapore

Arxiv: http://arxiv.org/abs/2411.18533v1

Refer to caption

Methods	Overall Results
	( Accuracy, Precision, Recall, F1 )
Resnet (Baseline)	79.17%, 79.56%, 78.99%, 78.87%
+ mean teacher	81.14%, 82.55%, 81.15%, 81.29%
+ SupConLoss	84.13%, 85.53%, 83.37%, 82.98%
+ mean teacher & SupConLoss	84.63%, 86.24%, 84.41%, 83.40%