MASTER: Multi-aspect non-local network for scene text recognition

作者：

Highlights：

• Multi-aspect non-local block enables the feature extracter to model global context.

• Different types of attention focus on different aspects of spatial feature dependencies.

• The inference speed is fast because of the proposed novel memory-cashed decoding mechanism.

• Our method achieves the best case-sensitive performance on COCO-text dataset.

摘要

•Multi-aspect non-local block enables the feature extracter to model global context.•Different types of attention focus on different aspects of spatial feature dependencies.•The inference speed is fast because of the proposed novel memory-cashed decoding mechanism.•Our method achieves the best case-sensitive performance on COCO-text dataset.

论文关键词：Scene text recognition,Transformer,Non-local network,Memory-cached mechanism

论文评审过程：Received 13 May 2020, Revised 16 September 2020, Accepted 30 March 2021, Available online 15 April 2021, Version of Record 18 April 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107980