用IMU数据实现鲁棒的跨作者手写识别，准确率超现有方法。
Robust and Efficient Writer-Independent IMU-Based Handwriting Recognition
- 结合CNN与双向LSTM，提升跨作者手写识别能力。
- 在公开数据集上达到7.37%字符错误率，句子级识别亦可行。
- 对不同年龄群体泛化性强，适合实际场景部署。
基于惯性测量单元（IMU）的手写识别（HWR）因书写风格差异大且数据集有限而面临挑战。以往方法在未见作者的书写上表现不佳，导致跨作者（WI）识别成为关键难题。本文提出一种新模型，采用卷积神经网络（CNN）编码器和双向长短期记忆网络（BiLSTM）解码器，显著提升对未见书写风格的鲁棒性。在公开的OnHW数据集和自建词级数据集的跨作者划分上，分别取得7.37%的字符错误率（CER）和9.44%的CER，以及15.12%和32.17%的词错误率（WER）。鲁棒性评估表明，模型在不同年龄组间具有优异泛化能力，知识迁移效果优于现有方法。在自建句级数据集上的实验进一步验证了全句识别潜力。消融实验显示，设计选择在性能与效率间达成良好平衡，为实际应用中更灵活、可扩展的HWR系统提供了支持。
原文摘要 · Abstract (English)
Handwriting recognition (HWR) using inertial measurement unit (IMU) data remains challenging due to variations in writing styles and the limited availability of datasets. Previous approaches often struggle with handwriting from unseen writers, making writer-independent (WI) recognition a crucial yet difficult problem. This paper presents a model designed to improve WI HWR on IMU data, using a CNN encoder and BiLSTM-based decoder. Our approach demonstrates strong robustness to unseen handwriting styles, outperforming existing methods on the WI splits of both the public OnHW dataset and our word-based dataset, achieving character error rates (CERs) of 7.37% and 9.44%, and word error rates (WERs) of 15.12% and 32.17%, respectively. Robustness evaluation shows that our model maintains superior performance across different age groups, with knowledge learned from one group generalizing better to another compared to other approaches. Evaluation on our sentence-based dataset further demonstrates the potential for recognizing full sentences. Through comprehensive ablation studies, we show that our design choices achieve a strong balance between performance and efficiency. These findings support the development of more adaptable and scalable HWR systems for real-world applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。