Self-Supervised Representation Learning for Online Handwriting Text Classification

TL;DR

This paper introduces a novel self-supervised learning method called Part of Stroke Masking (POSM) for online handwriting text classification, enabling effective representation learning for tasks like writer identification and gender classification.

Contribution

The study proposes a new pretext task, POSM, for self-supervised pretraining on online handwriting data in multiple languages, improving downstream classification performance.

Findings

Pretrained models outperform models trained from scratch.

State-of-the-art results in writer identification, gender, and handedness classification.

Effective representation learning demonstrated through intrinsic and extrinsic evaluations.

Abstract

Self-supervised learning offers an efficient way of extracting rich representations from various types of unlabeled data while avoiding the cost of annotating large-scale datasets. This is achievable by designing a pretext task to form pseudo labels with respect to the modality and domain of the data. Given the evolving applications of online handwritten texts, in this study, we propose the novel Part of Stroke Masking (POSM) as a pretext task for pretraining models to extract informative representations from the online handwriting of individuals in English and Chinese languages, along with two suggested pipelines for fine-tuning the pretrained models. To evaluate the quality of the extracted representations, we use both intrinsic and extrinsic evaluation methods. The pretrained models are fine-tuned to achieve state-of-the-art results in tasks such as writer identification, gender classification, and handedness classification, also highlighting the superiority of utilizing the pretrained models over the models trained from scratch.