On Study of the Reliable Fully Convolutional Networks with Tree Arranged Outputs (TAO-FCN) for Handwritten String Recognition

TL;DR

This paper presents a robust end-to-end handwritten string recognition system based on TAO-FCN, eliminating preprocessing and manual rules, emphasizing practicality over state-of-the-art accuracy.

Contribution

Introduces a fully convolutional network with tree-structured outputs for handwritten string recognition, simplifying the pipeline and enhancing robustness.

Findings

No preprocessing or manual rules needed

Easy to adapt with labeled data

Prioritizes usability and robustness

Abstract

The handwritten string recognition is still a challengeable task, though the powerful deep learning tools were introduced. In this paper, based on TAO-FCN, we proposed an end-to-end system for handwritten string recognition. Compared with the conventional methods, there is no preprocess nor manually designed rules employed. With enough labelled data, it is easy to apply the proposed method to different applications. Although the performance of the proposed method may not be comparable with the state-of-the-art approaches, it's usability and robustness are more meaningful for practical applications.