arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2304.13530cs.CVcs.AIcs.IR
收藏

从完整手写页面中提取键值信息

Key-value information extraction from full handwritten pages

TEKLIA(TEKLIA公司);LITIS(LITIS实验室);Normandy University(诺曼底大学)

Solène Tarride, Mélodie Boillet, Christopher Kermorvant

AI总结:

提出一种融合特征提取、手写识别与命名实体识别的 Transformer 模型,可直接处理完整手写页面并利用键值标注学习,在 IAM、ESPOSALLES 和 POPP 上均取得最优性能。

摘要:

我们提出了一种基于 Transformer 的方法,用于从数字化手写文档中提取信息。我们的方法在单一模型中结合了迄今为止由不同模型分别执行的各个步骤:特征提取、手写识别和命名实体识别。我们将这种集成方法与传统的两阶段方法(先执行手写识别,再执行命名实体识别)进行比较,并给出了行、段落和页面等不同层级的结果。我们的实验表明,基于注意力的模型在应用于完整页面时尤其具有吸引力,因为它们不需要任何预先的分割步骤。最后,我们证明它们能够从键值标注(即重要词及其对应命名实体的列表)中学习。我们在三个公开数据库(IAM、ESPOSALLES 和 POPP)上将模型与最先进方法进行比较,并在所有三个数据集上均超越了此前的性能。

英文摘要:

We propose a Transformer-based approach for information extraction from digitized handwritten documents. Our approach combines, in a single model, the different steps that were so far performed by separate models: feature extraction, handwriting recognition and named entity recognition. We compare this integrated approach with traditional two-stage methods that perform handwriting recognition before named entity recognition, and present results at different levels: line, paragraph, and page. Our experiments show that attention-based models are especially interesting when applied on full pages, as they do not require any prior segmentation step. Finally, we show that they are able to learn from key-value annotations: a list of important words with their corresponding named entities. We compare our models to state-of-the-art methods on three public databases (IAM, ESPOSALLES, and POPP) and outperform previous performances on all three datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑