TY - GEN
T1 - Character recognition in Japanese historical documents via adaptive multi-region model
AU - Wang, Yueyu
AU - Kamata, Sei Ichiro
N1 - Funding Information:
This work was partially supported by JSPS KAKENHI Grant Number 15K002248.
PY - 2019/2/12
Y1 - 2019/2/12
N2 - In this work, we introduce a novel model with an adaptive multi-region extraction network to grasp multi-aspect of discriminative features, because feature inside bounding box is insufficient for classification, and normal models are sensitive to inaccuracy of predicted bounding boxes. We use the new model to recognize Japanese from historical documents. This model can be trained end-to-end without any extra supervision. The resulting CNN-based representation has abundant of features, containing the contextual information together with center part information. These features are helpful and crucial for classification. Based on this model, we also propose a data augmentation method using both local and global data distortion to generate diversified samples in order to solve the problem of data imbalance. Experiments show that with the usage of our model, we get a better result in ancient Japanese dataset.
AB - In this work, we introduce a novel model with an adaptive multi-region extraction network to grasp multi-aspect of discriminative features, because feature inside bounding box is insufficient for classification, and normal models are sensitive to inaccuracy of predicted bounding boxes. We use the new model to recognize Japanese from historical documents. This model can be trained end-to-end without any extra supervision. The resulting CNN-based representation has abundant of features, containing the contextual information together with center part information. These features are helpful and crucial for classification. Based on this model, we also propose a data augmentation method using both local and global data distortion to generate diversified samples in order to solve the problem of data imbalance. Experiments show that with the usage of our model, we get a better result in ancient Japanese dataset.
UR - http://www.scopus.com/inward/record.url?scp=85063223673&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85063223673&partnerID=8YFLogxK
U2 - 10.1109/ICIEV.2018.8641033
DO - 10.1109/ICIEV.2018.8641033
M3 - Conference contribution
AN - SCOPUS:85063223673
T3 - 2018 Joint 7th International Conference on Informatics, Electronics and Vision and 2nd International Conference on Imaging, Vision and Pattern Recognition, ICIEV-IVPR 2018
SP - 404
EP - 409
BT - 2018 Joint 7th International Conference on Informatics, Electronics and Vision and 2nd International Conference on Imaging, Vision and Pattern Recognition, ICIEV-IVPR 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - Joint 7th International Conference on Informatics, Electronics and Vision and 2nd International Conference on Imaging, Vision and Pattern Recognition, ICIEV-IVPR 2018
Y2 - 25 June 2018 through 28 June 2018
ER -