MOoSE: Multi-Orientation Sharing Experts for Open-set Scene Text Recognition

TL;DR

This paper introduces MOoSE, a novel framework for open-set scene text recognition that effectively handles multi-orientation text by sharing expertise across different writing directions, addressing real-world challenges.

Contribution

The paper proposes MOoSE, a mixture-of-experts model for multi-orientation open-set text recognition, modeling diverse writing directions and novel characters in real-world scenes.

Findings

MOoSE effectively models multi-orientation text recognition.

The framework alleviates domain gaps between orientations.

Validation shows strong performance on benchmarks.

Abstract

Open-set text recognition, which aims to address both novel characters and previously seen ones, is one of the rising subtopics in the text recognition field. However, the current open-set text recognition solutions only focuses on horizontal text, which fail to model the real-life challenges posed by the variety of writing directions in real-world scene text. Multi-orientation text recognition, in general, faces challenges from the diverse image aspect ratios, significant imbalance in data amount, and domain gaps between orientations. In this work, we first propose a Multi-Oriented Open-Set Text Recognition task (MOOSTR) to model the challenges of both novel characters and writing direction variety. We then propose a Multi-Orientation Sharing Experts (MOoSE) framework as a strong baseline solution. MOoSE uses a mixture-of-experts scheme to alleviate the domain gaps between orientations, while exploiting common structural knowledge among experts to alleviate the data scarcity that some experts face. The proposed MOoSE framework is validated by ablative experiments, and also tested for feasibility on the existing open-set benchmark. Code, models, and documents are available at: https://github.com/lancercat/Moose/