EffViT-COVID: A dual-path network for COVID-19 percentage estimation

作者：

Highlights：

• Proposed a dual-path network for COVID-19 percentage estimation.

• An encoder mechanism is employed to effectively extract the rich features set.

• Extensive experiments are conducted to validate the proposed approach.

• Compared to existing methods, proposed approach achieves new state-of-the-art results.

摘要

•Proposed a dual-path network for COVID-19 percentage estimation.•An encoder mechanism is employed to effectively extract the rich features set.•Extensive experiments are conducted to validate the proposed approach.•Compared to existing methods, proposed approach achieves new state-of-the-art results.

论文关键词：COVID-19,Percentage estimation,EfficientNet-B7,Vision transformer,Huber loss,Deep network

论文评审过程：Received 14 July 2022, Revised 12 September 2022, Accepted 20 September 2022, Available online 3 October 2022, Version of Record 21 October 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118939