SAN: Structure-Aware Network for Complex and Long-tailed Chinese Text Recognition
Preprint
Abstract
In text recognition, complex glyphs and tail classes have always been factors affecting model performance. Specifically for Chinese text recognition, the lack of shape-awareness can lead to confusion among close complex characters. Since such characters are often tail classes that appear less frequently in the training-set, making it harder for the model to capture its shape information. Hence in this work, we propose a structure-aware network utilizing the hierarchical composition information to improve the recognition performance of complex characters. Implementation-wise, we first propose an auxiliary radical branch and integrate it into the base recognition network as a regularization term, which distills hierarchical composition information into the feature extractor. A Tree-Similarity-based weighting mechanism is then proposed to further utilize the depth information in the hierarchical representation. Experiments demonstrate that the proposed approach can significantly improve the performances of complex characters and tail characters, yielding a better overall performance. Code is available at https://github.com/Levi-ZJY/SAN.Keywords
All Related Versions
This publication has 19 references indexed in Scilit:
- Pho(SC)Net: An Approach Towards Zero-Shot Word Image Recognition in Historical Documents2021
- Joint architecture and knowledge distillation in CNN for Chinese text recognitionPattern Recognition, 2021
- Zero-shot Handwritten Chinese Character Recognition with hierarchical decomposition embeddingPattern Recognition, 2020
- Maximum Entropy Regularization and Chinese Text Recognition2020
- Towards Accurate Scene Text Recognition With Semantic Reasoning Networks2020
- SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition2020
- Writer-aware CNN for parsimonious HMM-based offline handwritten Chinese text recognitionPattern Recognition, 2020
- Radical aggregation network for few-shot offline handwritten Chinese character recognitionPattern Recognition Letters, 2019
- MORAN: A Multi-Object Rectified Attention Network for scene text recognitionPattern Recognition, 2019
- Building fast and compact convolutional neural networks for offline handwritten Chinese character recognitionPattern Recognition, 2017