摘要(英) |
In recent years, scene text analysis based on deep learning techniques draw
a lot of research attention. Text detection in natural scenes is an important step of
scene text analysis and most of the existing text detection designs are based on
string detection. However, a string may contain words of different languages so it
is not easy to mark the language to which the string belongs accurately. Scene text
recognition using string-level annotations need to consider the effect of irregular
orientations and require a lot of training data and training time. Conversely,
character-based recognition methodologies do not need to consider orientations,
which simplifies the training processes. Multilingual natural scene text
recognition may be benefited from the flexibility of selecting suitable recognition
models according to different language characteristics. In this research, we use a
high-resolution network architecture to label word regions and point out the
centers of characters, and also employ multiple channels for substring language
classification. Due to the lack of character-level annotations in real datasets, we
propose a weakly supervised learning approach for characters, enabling the
network to improve the detection of characters significantly. The performance of
multi-language recognition is verified by using individual classifiers after
detection or by performing language recognition at the same time. The feasibility
of the proposed design is verified by showing the character detection of different
languages, including Latin, Chinese, Japanese, and Korean, as examples. |
參考文獻 |
[1] Baoguang Shi, Mingkun Yang , Xinggang Wang, Pengyuan Lyu, Cong Yao , and Xiang Bai. “ ASTER: An Attentional Scene Text Recognizer with Flexible Rectification” CVPR 2016
[2] Canjie Luo, Lianwen Jin, Zenghui Sun. “A Multi-Object Rectified Attention Network for Scene Text Recognition.” CVPR2019
[3] Jaderg, Simonyan, Vedaldi, Zisserman. “Reading text in the wild with convolutional neural network”,IJCV2016
[4] Minghui Liao,Baoguang Shi,Xiang Bai, “Textboxes++ : A Single-shot Oriented Scene Text Detector.” TIP2018
[5] Kaiming He,Georgia Gkioxari,Piotr Dollar,Ross Girshick, “Mask R-CNN”, Facebook AI Research
[6] Yuliang Liu, Sheng Zhang, Lianwen Jim, Lele Xie, Yaqiang Wu, Zhepeng Wang. “Omnidirectional Scene Text Detection with Sequential-free Box Discretization.” Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, 2019
[7] X. Zhou,C. Yao,H.Wen,Y. Wang,S. Zhou,W.He, J.Liang “EAST: An efficient and Accurate Scene Text Detector.”,CVPR 2017
[8] Jonathan Long, Evan Shelhamer, Trevor Darrell. “Fully Convolution Networks for Semantic Segmentation.” CVPR2015
[9] Minghui Liao, Zhaoyi Wan,Cong Yao,Kai Chan,Xiang Bai, “Real-time Scene Text Detection with Differentiable Binarization.” AAAI2020
[10] Jian Ye, Zhe Chen, Juhua Liu, Bo Du. “TextFuseNet: Scene Text Detection with Richer Fused Features.” IJCAI2020
[11] Youngmin Baek, Bado Lee, Dongyoon Han, Sangdoo Tun, Hwalsuk Lee. “Character Region Awareness For Text Detection.” CVPR2019
[12] Linjie Xing, Zhi Tian , Weilin Huang and Matthew R. Scott. “Convolutional Character Networks.” ICCV 2019
[13] Olaf Ronneberger, Philipp Fischer, Thomas Brox. “U-Net: Convolutional Networks for Biomedical Image Segmentation.” MICCAI2015
[14] Zhi-Hua Zhou《A Brief Introduction to Weakly Supervised Learning》journal National Science Review 2018.1
[15] Shangbang Long,Xin He,Cong Yao, “Scene Text Detection and Recognition: The Deep Learning Era”, IJCV2021
[16] Ankush Gupta, Andrea Vedakdi and Andrew Zisserman. “SynthText in the Wild Dataset.” CVPR2016
[17] Max Jaderberg, Karen Simonyan, Andrea Vedaldi and Andrew Zisserman. “Reading Text in the Wild with Convolution Neural Networks.” International journal of computer vision 2016
[18] Prasun Roy, Saumik Bhattacherya, Subhankar Ghosh, Umapada Pal. “STEFANN: Scene Text Editor using Font Adaptive Neural Network.” CVPR’20
[19] Shangbang Long, Cong Yao “UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World” CVPR2020
[20] PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing, ACM Transactions on Graphics journal 2009
[21] Ke Sun,Bin Xiao,Dong Liu,Jingdong Wang. “Deep High-Resolution Representation Learning for Human Pose Estimation”, CVPR2019
[22] Mingxing Tan,Quoc V. Le. “EfficientNet: Rethinking Model Scaling for Convolutioin Neural Networks.”
[23] Nibal Nayef, Fei Yin, Imen Bizid, Hyunsoo Choi, Yuan Feng, “Icdar2017 Robust Reading Challenge on Multi-lingual Scene Text Detection and Script Identification”, 14th Iapr International Conference on Document Analysis and Recognition, 2017.
[24] Chee Kheng Chng,Chee Seng Chan. “Total-Text:A Comprehensive Dataset for Scene Text Detection and Recognition.” ICDAR2017 |