Tokenization vs. Augmentation: A Systematic Study of Writer Variance in IMU-Based Online Handwriting Recognition

2026-02-25

arXiv.org

AI Summary

This study addresses the generalization challenges in inertial measurement unit (IMU)-based online handwriting recognition caused by imbalanced character distributions and inter-writer variability. The authors systematically compare two complementary strategies: subword segmentation using Bigram modeling to enhance cross-writer generalization, and concatenative data augmentation to alleviate data sparsity for individual writers. Integrated within an end-to-end sequence recognition framework that explicitly models IMU signals, both approaches demonstrate synergistic benefits. On the OnHW-Words500 dataset, the proposed methods achieve a word error rate (WER) of 12.99% in cross-writer scenarios. In writer-dependent settings, they reduce the character error rate (CER) and WER by 34.5% and 25.4%, respectively.

Abstract

Inertial measurement unit-based online handwriting recognition enables the recognition of input signals collected across different writing surfaces but remains challenged by uneven character distributions and inter-writer variability. In this work, we systematically investigate two strategies to address these issues: sub-word tokenization and concatenation-based data augmentation. Our experiments on the OnHW-Words500 dataset reveal a clear dichotomy between handling inter-writer and intra-writer variance. On the writer-independent split, structural abstraction via Bigram tokenization significantly improves performance to unseen writing styles, reducing the word error rate (WER) from 15.40% to 12.99%. In contrast, on the writer-dependent split, tokenization degrades performance due to vocabulary distribution shifts between the training and validation sets. Instead, our proposed concatenation-based data augmentation acts as a powerful regularizer, reducing the character error rate by 34.5% and the WER by 25.4%. Further analysis shows that short, low-level tokens benefit model performance and that concatenation-based data augmentation performance gain surpasses those achieved by proportionally extended training. These findings reveal a clear variance-dependent effect: sub-word tokenization primarily mitigates inter-writer stylistic variability, whereas concatenation-based data augmentation effectively compensates for intra-writer distributional sparsity.

Problem

• handwriting recognition
• inter-writer variability
• intra-writer variance
• IMU-based
• tokenization

Innovation

• subword tokenization
• concatenation-based data augmentation
• inter-writer variance
• intra-writer variance
• IMU-based handwriting recognition