FMD-Yolo: An efficient face mask detection method for COVID-19 prevention and control in public

作者：

Highlights：

• Hierarchical and fine-grained framework is recommended for face mask detection.

• Task-relevant prior knowledge benefits model training with stronger robustness.

• Self-attention mechanism facilitates sufficient fusion among features in different levels.

• Generalization ability of a model is hard to guarantee while applying in different scenarios.

摘要

Highlights•Hierarchical and fine-grained framework is recommended for face mask detection.•Task-relevant prior knowledge benefits model training with stronger robustness.•Self-attention mechanism facilitates sufficient fusion among features in different levels.•Generalization ability of a model is hard to guarantee while applying in different scenarios.

论文关键词：Face mask detection,COVID-19,Improved YoloV3 algorithm,Feature extraction and fusion

论文评审过程：Received 31 August 2021, Revised 8 October 2021, Accepted 19 November 2021, Available online 25 November 2021, Version of Record 2 December 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104341