講演抄録/キーワード |
講演名 |
2018-09-21 13:40
[ショートペーパー]Stroke Recovery of Handwritten Chinese Character using Fully Convolutional Networks ○Wang Yu Jung・Motoharu Sonogashira(Kyoto Univ.)・Atsushi Hashimoto(Omron SINICX)・Masaaki Iiyama(Kyoto Univ.) PRMU2018-62 IBISML2018-39 |
抄録 |
(和) |
In this paper, we propose a method to recover strokes from offline handwritten Chinese characters. The proposed method employs multitask fully convolutional networks (FCN) to estimate writing order and direction of strokes from offline character images. Online data in CASIA database is used for training. Since the FCN produces discontinuous strokes, our method refines the estimation of the writing order by using a graph cut (GC). Experimental results with test dataset of CASIA-OLHWDB1.0 show the effectiveness of our method. |
(英) |
In this paper, we propose a method to recover strokes from offline handwritten Chinese characters. The proposed method employs multitask fully convolutional networks (FCN) to estimate writing order and direction of strokes from offline character images. Online data in CASIA database is used for training. Since the FCN produces discontinuous strokes, our method refines the estimation of the writing order by using a graph cut (GC). Experimental results with test dataset of CASIA-OLHWDB1.0 show the effectiveness of our method. |
キーワード |
(和) |
Handwriting Trajectory Recovery / Semantic Segmentation / Convolutional Networks / Deep Learning / / / / |
(英) |
Handwriting Trajectory Recovery / Semantic Segmentation / Convolutional Networks / Deep Learning / / / / |
文献情報 |
信学技報, vol. 118, no. 219, PRMU2018-62, pp. 157-162, 2018年9月. |
資料番号 |
PRMU2018-62 |
発行日 |
2018-09-13 (PRMU, IBISML) |
ISSN |
Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
PRMU2018-62 IBISML2018-39 |