Reciprocal Feature Learning via Explicit and Implicit Tasks in Scene Text Recognition

TL;DR

This paper introduces a reciprocal feature learning framework for scene text recognition that leverages both explicit sequential recognition and an implicit character counting task to improve accuracy and robustness.

Contribution

It proposes a novel two-branch reciprocal learning framework that utilizes implicit character counting as an auxiliary task to enhance scene text recognition.

Findings

Outperforms existing methods on 7 benchmarks.

Effectively integrates implicit character counting with recognition.

Demonstrates robustness across different network architectures.

Abstract

Text recognition is a popular topic for its broad applications. In this work, we excavate the implicit task, character counting within the traditional text recognition, without additional labor annotation cost. The implicit task plays as an auxiliary branch for complementing the sequential recognition. We design a two-branch reciprocal feature learning framework in order to adequately utilize the features from both the tasks. Through exploiting the complementary effect between explicit and implicit tasks, the feature is reliably enhanced. Extensive experiments on 7 benchmarks show the advantages of the proposed methods in both text recognition and the new-built character counting tasks. In addition, it is convenient yet effective to equip with variable networks and tasks. We offer abundant ablation studies, generalizing experiments with deeper understanding on the tasks. Code is available.