Two-Stage Fully Convolutional Networks for Stroke Recovery of Handwritten Chinese Character
- 72 Downloads
In this paper, we propose a method to recover strokes from offline handwritten Chinese characters. The proposed method employs a fully convolutional network (FCN) to estimate the writing order of connected components in offline Chinese character images and a multi-task FCN to estimate the writing order and directions of strokes in each connected component. Online dataset CASIA-OLHWDB1.0 from the CASIA database is hired as the training set. Because the network produces discontinuous strokes, we refine the estimated writing orders using a graph cut (GC), in which the estimated directions are used for calculation of smoothness term. Experimental results with test dataset of CASIA-OLHWDB1.0tst demonstrate the effectiveness of our method.
KeywordsHandwriting trajectory recovery Semantic segmentation Fully convolutional networks Graphcut
This work was supported by JSPS KAKENHI Grant Number 17H06288.
- 4.Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, pp. 3431–3440 (2015)Google Scholar
- 6.Doermann, D.S., Rosenfeld, A.: Recovery of temporal information from static images of handwriting. In: Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Champaign, IL, USA, pp. 162–168 (1992)Google Scholar
- 10.Zhao, B., Yang, M., Tao, J.: Drawing order recovery for handwriting Chinese characters. In: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, United Kingdom, pp. 3227–3231 (2019)Google Scholar
- 11.Lallican, P.M., Viard-Gaudin, C.: A Kalman approach for stroke order recovering from off-line handwriting. In: Proceedings of the Fourth International Conference on Document Analysis and Recognition, Ulm, Germany, vol. 2, pp. 519–522 (1997)Google Scholar
- 14.Kato, Y., Yasuhara, M.: Recovery of drawing order from scanned images of multi-stroke handwriting. In: Fifth International Conference on Document Analysis and Recognition, pp. 261–264 (1999)Google Scholar
- 15.Qiao, Y., Yasuhara, M.: Recover writing trajectory from multiple stroked image using bidirectional dynamic search. In: 18th International Conference on Pattern Recognition (ICPR 2006), Hong Kong, pp. 970–973 (2006)Google Scholar