Quantitative Analyses on Effects from Constraints in Air-Writing

Songbin XU  Yang XUE  Yuqing CHEN  

IEICE TRANSACTIONS on Information and Systems   Vol.E102-D   No.4   pp.867-870
Publication Date: 2019/04/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2018EDL8042
Type of Manuscript: LETTER
Category: Artificial Intelligence, Data Mining
inertial sensor,  air-writing recognition,  quantitative analyses,  LSTM-RNN,  

Full Text: FreePDF(2.4MB)

Very few existing works about inertial sensor based air-writing focused on writing constraints' effects on recognition performance. We proposed a LSTM-based system and made several quantitative analyses under different constraints settings against CHMM, DTW-AP and CNN. The proposed system shows its advantages in accuracy, real-time performance and flexibility.