Gunma University · Faculty of Informatics

Research keywordshistorical cursive OCR・document analysis・handwriting recognition・semi-supervised learning・CRNN・line image datasets

This is a public-data preview. Personalized fit, contact angles, and saved workflows require sign-in.

• Generation of a Large-Scale Line Image Dataset with Ground Truth Texts from Page-Level Autograph Documents, Ayumu Nagai, 2021年12月, 28th International Conference on Neural Information Processing (ICONIP), 354, 3662021 · 論文
• Recognizing Japanese Historical Cursive with Pseudo-Labeling-aided CRNN as an Application of Semi-Supervised Learning to Sequence Labeling, Ayumu Nagai, 2020年, 17th International Conference on Frontiers in Handwriting Recognition (ICFHR), 97, 102, 研究論文（国際会議プロシーディングス）2020 · 論文
• On the Improvement of Recognizing Single-line Strings of Japanese Historical Cursive, Ayumu Nagai, 2019年, 2019 International Conference on Document Analysis and Recognition (ICDAR), 621, 628, 研究論文（国際会議プロシーディングス）2019 · 論文
• 畳み込みネットワークによる仮名３文字のくずし字の文字認識, 長井 歩, 2017年12月, 人文科学とコンピュータ論文集, 213-2182017 · 論文

• Official profile