Robust and Efficient Writer-Independent IMU-Based Handwriting Recognition

TL;DR

This paper introduces a CNN-BiLSTM model for IMU-based handwriting recognition that is highly robust to unseen writers and styles, outperforming existing methods on multiple datasets and demonstrating strong generalization capabilities.

Contribution

The paper proposes a novel CNN encoder and BiLSTM decoder architecture that significantly improves writer-independent IMU handwriting recognition performance and robustness.

Findings

Achieved CER of 7.37% and WER of 15.12% on public dataset

Maintains accuracy across different age groups

Demonstrates effective sentence recognition in full sentences

Abstract

Online handwriting recognition (HWR) using data from inertial measurement units (IMUs) remains challenging due to variations in writing styles and the limited availability of annotated datasets. Previous approaches often struggle with handwriting from unseen writers, making writer-independent (WI) recognition a crucial yet difficult problem. This paper presents an HWR model designed to improve WI HWR on IMU data, using a CNN encoder and a 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 accuracy across different age groups, and knowledge learned from one group generalizes better to another. Evaluation on our sentence-based dataset further demonstrates its potential in recognizing full sentences. Through comprehensive ablation studies, we show that our design choices lead to a strong balance between performance and efficiency. These findings support the development of more adaptable and scalable HWR systems for real-world applications.