Towards Accurate Scene Text Recognition with Semantic Reasoning Networks

Abstract

Scene text image contains two levels of contents: visual texture and semantic information. Although the previous scene text recognition methods have made great progress over the past few years, the research on mining semantic information to assist text recognition attracts less attention, only RNN-like structures are explored to implicitly model semantic information. However, we observe that RNN based methods have some obvious shortcomings, such as time-dependent decoding manner and one-way serial transmission of semantic context, which greatly limit the help of semantic information and the computation efficiency. To mitigate these limitations, we propose a novel end-to-end trainable framework named semantic reasoning network (SRN) for accurate scene text recognition, where a global semantic reasoning module (GSRM) is introduced to capture global semantic context through multi-way parallel transmission. The state-of-the-art results on 7 public benchmarks, including regular text, irregular text and non-Latin long text, verify the effectiveness and robustness of the proposed method. In addition, the speed of SRN has significant advantages over the RNN based methods, demonstrating its value in practical use.

arXiv commentAccepted to CVPR2020

References in corpus (11)

• Convolutional Sequence to Sequence Learning
• A Structured Self-attentive Sentence Embedding
• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
• Focusing Attention: Towards Accurate Text Recognition in Natural Images
• Neural Machine Translation in Linear Time
• Synthetic Data for Text Localisation in Natural Images
• Deep Structured Output Learning for Unconstrained Text Recognition
• Scene Text Recognition with Sliding Convolutional Character Models
• 2D Attentional Irregular Scene Text Recognizer
• ICDAR 2015 Text Reading in the Wild Competition
• An End-to-end Video Text Detector with Online Tracking

Cited by in corpus (12)

• Text Recognition in the Wild: A Survey
• MMOCR: A Comprehensive Toolbox for Text Detection, Recognition and Understanding
• Visual-Semantic Transformer for Scene Text Recognition
• TRIG: Transformer-Based Text Recognizer with Initial Embedding Guidance
• Hamming OCR: A Locality Sensitive Hashing Neural Network for Scene Text Recognition
• StrucTexT: Structured Text Understanding with Multi-Modal Transformers
• What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels
• Primitive Representation Learning for Scene Text Recognition
• SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition
• Parallel Scale-wise Attention Network for Effective Scene Text Recognition
• PIMNet: A Parallel, Iterative and Mimicking Network for Scene Text Recognition
• Utilizing Resource-Rich Language Datasets for End-to-End Scene Text Recognition in Resource-Poor Languages