Towards Accurate Scene Text Recognition With Semantic Reasoning Networks
By Deli Yu, Xuan Li, Chengquan Zhang, Junyu Han, Jingtuo Liu, Errui Ding. 2020
344 citations recorded by the source corpus.
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.
Subfields: Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Zhe Gan
- Zicheng Liu
- Lijuan Wang
- Luc Van Gool
- Dacheng Tao
- Cordelia Schmid
- Jianfeng Gao
- Jitendra Malik
Citations in the corpus
- GIT: A Generative Image-to-text Transformer for Vision and Language (2022, background): “[121] Deli Yu, Xuan Li, Chengquan Zhang, Tao Liu, Junyu Han, Jingtuo Liu, and Errui Ding.”
- Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition (2021, neutral): “SRN (Yu et al. 2020) uses transformer-based models where global semantic information as well as long-range word dependency is modelled by self-attention.”
- PP-OCR: A Practical Ultra Lightweight OCR System (2020, background): “Data Augmentation This paper (Yu et al. 2020) shows some image processing operations to train a text recognizer, such as rotation, perspective distortion, motion blur and Gaussian noise.”
- MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining (2022, methodology): “In [12, 20, 30, 36, 44, 45, 60], the recognition model is pre-trained on synthesized data using supervised learning.”
- TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models (2021, background): “(Yu et al. 2020) proposed semantic reasoning networks to replace RNN-like structures for more accurate text recognition.”