Y
Yejun Tang
Researcher at Tsinghua University
Publications - 5
Citations - 84
Yejun Tang is an academic researcher from Tsinghua University. The author has contributed to research in topics: Convolutional neural network & Intelligent character recognition. The author has an hindex of 5, co-authored 5 publications receiving 58 citations.
Papers
More filters
Proceedings ArticleDOI
CNN Based Transfer Learning for Historical Chinese Character Recognition
TL;DR: A transfer learning method based on Convolutional Neural Network for historical Chinese character recognition and several experiments regarding essential factors of the CNNbased transfer learn method are conducted, showing that the proposed method is effective.
Proceedings ArticleDOI
TH-GAN: Generative Adversarial Network Based Transfer Learning for Historical Chinese Character Recognition
TL;DR: A weighted mean squared error criterion by incorporating both the edge and the skeleton information in the ground truth image is proposed as the weighted pixel loss in the generator to preserve the complex glyph structure of a historical Chinese character.
Proceedings ArticleDOI
Semi-Supervised Transfer Learning for Convolutional Neural Network Based Chinese Character Recognition
TL;DR: Experimental results on practical Chinese character transfer learning tasks, such as Dunhuang historical Chinese character recognition, indicate that the proposed semi-supervised transfer learning method can significantly improve recognition accuracy in the target domain.
Proceedings ArticleDOI
On-line Handwritten Mathematical Expression Recognition Method Based on Statistical and Semantic Analysis
Yang Hu,Liangrui Peng,Yejun Tang +2 more
TL;DR: A novel framework to analyse HME layout and semantic information is presented, namely symbol segmentation, symbol recognition and semantic relationship analysis, which includes a ternary tree to store the ranked symbols through calculating the operator priorities.
Patent
Semi-supervised-transfer-learning character recognition method and system based on convolutional neural network
TL;DR: In this paper, a semi-supervised-transfer learning character recognition method based on a convolutional neural network is proposed. But the method is limited to the target domain and has no class labels.