Out of Length Text Recognition with Sub-String Matching

TL;DR

This paper introduces a new long text recognition benchmark and a novel sub-string matching method, SMTR, that effectively recognizes arbitrarily long text by iteratively matching sub-strings, even trained only on short text datasets.

Contribution

The paper presents the first Long Text Benchmark (LTB) and a novel sub-string matching approach, SMTR, for out-of-length text recognition, enabling effective long text recognition from short text training data.

Findings

SMTR outperforms existing methods on short text benchmarks.

SMTR achieves superior performance on the new LTB.

Regularization and inference strategies improve recognition accuracy.

Abstract

Scene Text Recognition (STR) methods have demonstrated robust performance in word-level text recognition. However, in real applications the text image is sometimes long due to detected with multiple horizontal words. It triggers the requirement to build long text recognition models from readily available short (i.e., word-level) text datasets, which has been less studied previously. In this paper, we term this task Out of Length (OOL) text recognition. We establish the first Long Text Benchmark (LTB) to facilitate the assessment of different methods in long text recognition. Meanwhile, we propose a novel method called OOL Text Recognition with sub-String Matching (SMTR). SMTR comprises two cross-attention-based modules: one encodes a sub-string containing multiple characters into next and previous queries, and the other employs the queries to attend to the image features, matching the sub-string and simultaneously recognizing its next and previous character. SMTR can recognize text of arbitrary length by iterating the process above. To avoid being trapped in recognizing highly similar sub-strings, we introduce a regularization training to compel SMTR to effectively discover subtle differences between similar sub-strings for precise matching. In addition, we propose an inference augmentation strategy to alleviate confusion caused by identical sub-strings in the same text and improve the overall recognition efficiency. Extensive experimental results reveal that SMTR, even when trained exclusively on short text, outperforms existing methods in public short text benchmarks and exhibits a clear advantage on LTB. Code: https://github.com/Topdu/OpenOCR.