Revisiting Classification Perspective on Scene Text Recognition

TL;DR

This paper proposes a simple, classification-based approach to scene text recognition that achieves competitive performance with less complex implementation and annotation requirements.

Contribution

It introduces the CSTR model, demonstrating that classification perspective can match other methods in accuracy while simplifying the pipeline.

Findings

CSTR achieves near state-of-the-art results on six benchmarks.

The classification approach simplifies implementation and deployment.

CSTR performs well on both regular and irregular text datasets.

Abstract

The prevalent perspectives of scene text recognition are from sequence to sequence (seq2seq) and segmentation. Nevertheless, the former is composed of many components which makes implementation and deployment complicated, while the latter requires character level annotations that is expensive. In this paper, we revisit classification perspective that models scene text recognition as an image classification problem. Classification perspective has a simple pipeline and only needs word level annotations. We revive classification perspective by devising a scene text recognition model named as CSTR, which performs as well as methods from other perspectives. The CSTR model consists of CPNet (classification perspective network) and SPPN (separated conv with global average pooling prediction network). CSTR is as simple as image classification model like ResNet \cite{he2016deep} which makes it easy to implement and deploy. We demonstrate the effectiveness of the classification perspective on scene text recognition with extensive experiments. Futhermore, CSTR achieves nearly state-of-the-art performance on six public benchmarks including regular text, irregular text. The code will be available at https://github.com/Media-Smart/vedastr.