Presentation | 2022-03-11 Self-supervised Contrastive Learning Using Triplet Loss for Offline Recognition of Handwritten Chinese Text lines Trung Tan Ngo, Hung Tuan Nguyen, Masaki Nakagawa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose a framework for contrastive learning of visual representations using online triplet loss and apply it for offline recognition of handwritten Chinese text lines. In this framework, the visual encoder model is trained with unlabeled text line images, then finetuned on ones with labels. As far as we know, it is the first approach that uses self-supervised contrastive learning for Chinese text line recognition. We apply the CRNN model to recognize text line images. At first, only the CNN part is trained in the proposed framework, and then it is used as the initial weight for the CRNN model when finetuned. In the experiments, we evaluated the performance of the proposed framework on the CASIA dataset. The results show that the text line recognizer trained with the self-supervised pre-trained encoder has outperformed the one without the pre-trained model. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Machine learning / contrastive learning / offline handwriting recognition / deep neural network / CRNN |
Paper # | PRMU2021-78 |
Date of Issue | 2022-03-03 (PRMU) |
Conference Information | |
Committee | PRMU / IPSJ-CVIM |
---|---|
Conference Date | 2022/3/10(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Differentiable rendering |
Chair | Seiichi Uchida(Kyushu Univ.) |
Vice Chair | Masakazu Iwamura(Osaka Pref. Univ.) / Mitsuru Anpai(Denso IT Lab.) |
Secretary | Masakazu Iwamura(NTT) / Mitsuru Anpai(Tottori Univ.) |
Assistant | Kouta Yamaguchi(CyberAgent) / Yusuke Matsui(Univ. of Tokyo) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Special Interest Group on Computer Vision and Image Media |
---|---|
Language | ENG |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Self-supervised Contrastive Learning Using Triplet Loss for Offline Recognition of Handwritten Chinese Text lines |
Sub Title (in English) | |
Keyword(1) | Machine learning |
Keyword(2) | contrastive learning |
Keyword(3) | offline handwriting recognition |
Keyword(4) | deep neural network |
Keyword(5) | CRNN |
1st Author's Name | Trung Tan Ngo |
1st Author's Affiliation | Tokyo university of Agriculture and Technology(TUAT) |
2nd Author's Name | Hung Tuan Nguyen |
2nd Author's Affiliation | Tokyo university of Agriculture and Technology(TUAT) |
3rd Author's Name | Masaki Nakagawa |
3rd Author's Affiliation | Tokyo university of Agriculture and Technology(TUAT) |
Date | 2022-03-11 |
Paper # | PRMU2021-78 |
Volume (vol) | vol.121 |
Number (no) | PRMU-427 |
Page | pp.pp.115-120(PRMU), |
#Pages | 6 |
Date of Issue | 2022-03-03 (PRMU) |