ACMU-nets: Attention cascading modular U-nets incorporating squeeze and excitation blocks

研究成果: Chapter in Book/Report/Conference proceedingConference contribution

抄録

In document analysis research, image-to-image conversion models such as a U-Net have been shown significant performance. Recently, cascaded U-Nets research is suggested for solving complex document analysis studies. However, improving performance by adding U-Net modules requires using too many parameters in cascaded U-Nets. Therefore, in this paper, we propose a method for enhancing the performance of cascaded U-Nets. We suggest a novel document image binarization method by utilizing Cascading Modular U-Nets (CMU-Nets) and Squeeze and Excitation blocks (SE-blocks). Through verification experiments, we point out the problems caused by the use of SE-blocks in existing CMU-Nets and suggest how to use SE-blocks in CMU-Nets. We use the Document Image Binarization (DIBCO) 2017 dataset to evaluate the proposed model.

本文言語英語
ホスト出版物のタイトルDocument Analysis Systems - 14th IAPR International Workshop, DAS 2020, Proceedings
編集者Xiang Bai, Dimosthenis Karatzas, Daniel Lopresti
出版社Springer
ページ118-130
ページ数13
ISBN(印刷版)9783030570576
DOI
出版ステータス出版済み - 2020
イベント14th IAPR International Workshop on Document Analysis Systems, DAS 2020 - Wuhan, 中国
継続期間: 7 26 20207 29 2020

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
12116 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

会議

会議14th IAPR International Workshop on Document Analysis Systems, DAS 2020
Country中国
CityWuhan
Period7/26/207/29/20

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

フィンガープリント 「ACMU-nets: Attention cascading modular U-nets incorporating squeeze and excitation blocks」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル