On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention

10/10/2019
by   Junyeop Lee, et al.
13

Scene text recognition (STR) is the task of recognizing character sequences in natural scenes. While there have been great advances in STR methods, current methods still fail to recognize texts in arbitrary shapes, such as heavily curved or rotated texts, which are abundant in daily life (e.g. restaurant signs, product labels, company logos, etc). This paper introduces a novel architecture to recognizing texts of arbitrary shapes, named Self-Attention Text Recognition Network (SATRN), which is inspired by the Transformer. SATRN utilizes the self-attention mechanism to describe two-dimensional (2D) spatial dependencies of characters in a scene text image. Exploiting the full-graph propagation of self-attention, SATRN can recognize texts with arbitrary arrangements and large inter-character spacing. As a result, SATRN outperforms existing STR models by a large margin of 5.7 pp on average in "irregular text" benchmarks. We provide empirical analyses that illustrate the inner mechanisms and the extent to which the model is applicable (e.g. rotated and multi-line text). We will open-source the code.

READ FULL TEXT

page 1

page 6

page 8

research
11/12/2017

Arbitrarily-Oriented Text Recognition

Recognizing text from natural images is still a hot research topic in co...
research
03/25/2020

SCATTER: Selective Context Attentional Scene Text Recognizer

Scene Text Recognition (STR), the task of recognizing text against compl...
research
03/18/2020

Scene Text Recognition via Transformer

Scene text recognition with arbitrary shape is very challenging due to l...
research
01/01/2022

SAFL: A Self-Attention Scene Text Recognizer with Focal Loss

In the last decades, scene text recognition has gained worldwide attenti...
research
07/11/2022

COO: Comic Onomatopoeia Dataset for Recognizing Arbitrary or Truncated Texts

Recognizing irregular texts has been a challenging topic in text recogni...
research
09/16/2020

Exploring Font-independent Features for Scene Text Recognition

Scene text recognition (STR) has been extensively studied in last few ye...
research
04/15/2021

Rethinking Text Line Recognition Models

In this paper, we study the problem of text line recognition. Unlike mos...

Please sign up or login with your details

Forgot password? Click here to reset