Learning Spatial-Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition

TL;DR

This paper introduces a novel fully convolutional recurrent network that leverages path signature features and an implicit language model to improve online handwritten Chinese text recognition, achieving state-of-the-art accuracy.

Contribution

It proposes a new multi-spatial-context network and an implicit language model that effectively handle segmentation and semantic context in Chinese handwriting recognition.

Findings

Achieved 97.10% and 97.15% accuracy on two benchmarks.

Outperformed all previous methods in recognition accuracy.

Successfully integrated semantic context for improved recognition.

Abstract

Online handwritten Chinese text recognition (OHCTR) is a challenging problem as it involves a large-scale character set, ambiguous segmentation, and variable-length input sequences. In this paper, we exploit the outstanding capability of path signature to translate online pen-tip trajectories into informative signature feature maps using a sliding window-based method, successfully capturing the analytic and geometric properties of pen strokes with strong local invariance and robustness. A multi-spatial-context fully convolutional recurrent network (MCFCRN) is proposed to exploit the multiple spatial contexts from the signature feature maps and generate a prediction sequence while completely avoiding the difficult segmentation problem. Furthermore, an implicit language model is developed to make predictions based on semantic context within a predicting feature sequence, providing a new perspective for incorporating lexicon constraints and prior knowledge about a certain language in the recognition procedure. Experiments on two standard benchmarks, Dataset-CASIA and Dataset-ICDAR, yielded outstanding results, with correct rates of 97.10% and 97.15%, respectively, which are significantly better than the best result reported thus far in the literature.