Candidate fusion: Integrating language modelling into a sequence-to-sequence handwritten word recognition architecture

作者：

Highlights：

• A novel integrated language model for handwriting word recognition.

• A sequence-to-sequence approach for handwritten word recognition..

• Candidate fusion incorporates language statistics and commonly produced errors.

• We achieve state-of-the-art performances challenging datasets.

摘要

•A novel integrated language model for handwriting word recognition.•A sequence-to-sequence approach for handwritten word recognition..•Candidate fusion incorporates language statistics and commonly produced errors.•We achieve state-of-the-art performances challenging datasets.

论文关键词：Handwritten word recognition,Sequence-to-sequence models,Language model,Candidate fusion

论文评审过程：Received 21 December 2019, Revised 10 October 2020, Accepted 7 December 2020, Available online 17 December 2020, Version of Record 25 December 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107790