如圖4所示,(a)展示了一系列的文本組件,每個文本組件D由(x,y,h,w,sinθ,cosθ)組成。其中h爲文本組件的高度,由(c )圖中的h1和h2兩部分組成。w則是根據h的大小確定的。
(b)中展示了文本組建的中心域,爲了確定文本中心域(text center region 記爲TCR)與文本組件的方向,本文采用了[17]中的方法來計算文本域的head和tail,如圖4(a)中黑色箭頭所示。
[17] Shangbang Long, Jiaqiang Ruan, Wenjie Zhang, Xin He, Wenhao Wu, and Cong Yao. Textsnake: A flexible repre-
sentation for detecting text of arbitrary shapes. In ECCV,
pages 19–35, 2018. 1, 2, 3, 4, 7, 8
[1] Youngmin Baek, Bado Lee, Dongyoon Han, Sangdoo Yun, and Hwalsuk Lee. Character region awareness for text de- tection. In CVPR, pages 9365–9374, 2019. 1, 2, 3, 7, 8
[11] Minghui Liao, Zhen Zhu, Baoguang Shi, Gui-Song Xia, and Xiang Bai. Rotation-sensitive regression for oriented scene text detection. In CVPR, pages 5909–5918, 2018. 2
[13] Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott E. Reed, Cheng-Yang Fu, and Alexander C. Berg. SSD: Single shot multibox detector. In ECCV, pages 21–37, 2016. 2
[10] Minghui Liao, Baoguang Shi, and Xiang Bai. Textboxes++: A single-shot oriented scene text detector. IEEE Transac- tions on Image Processing, 27(8):3676–3690, 2018. 2, 8
[42] Xinyu Zhou, C.Yao, He Wen, Yuzhi Wang, Shuchang Zhou, Weiran He, and Jiajun Liang. EAST: An efficient and accu- rate scene text detector. In CVPR, pages 2642–2651, 2017. 1, 2, 8
[3] Dan Deng, Haifeng Liu, Xuelong Li, and Deng Cai. Pix- elLink: Detecting scene text via instance segmentation. In AAAI, pages 6773–6780, 2018. 2, 8
[30] Wenhai Wang, Enze Xie, Xiang Li, Wenbo Hou, Tong Lu, Gang Yu, and Shuai Shao. Shape robust text detection with progressive scale expansion network. In CVPR, pages 9336– 9345, 2019. 2, 7, 8
[28] ZhuotaoTian,MichelleShu,PengyuanLyu,RuiyuLi,Chao Zhou, Xiaoyong Shen, and Jiaya Jia. Learning shape-aware embedding for scene text detection. In CVPR, pages 4234– 4243, 2019. 2, 7, 8
[34] Yongchao Xu, Yukang Wang, Wei Zhou, Yongpan Wang, Zhibo Yang, and Xiang Bai. Textfield: Learning a deep di- rection field for irregular scene text detection. IEEE Trans. Image Processing, 28(11):5566–5579, 2019. 2, 7, 8
[27] ZhiTian,WeilinHuang,TongHe,PanHe,andYuQiao.De- tecting text in natural image with connectionist text proposal network. In ECCV, pages 56–72, 2016. 1, 2
[21] Baoguang Shi, Xiang Bai, and Serge J. Belongie. Detect- ing oriented text in natural images by linking segments. In CVPR, pages 3482–3490, 2017. 1, 2, 7, 8
[4] Wei Feng, Wenhao He, Fei Yin, Xu-Yao Zhang, and Cheng- Lin Liu. Textdragon: An end-to-end framework for arbitrary shaped text spotting. In ICCV, pages 9075–9084, 2019. 2, 7
[33] Zhongdao Wang, Liang Zheng, Yali Li, and Shengjin Wang. Linkage based face clustering via graph convolution net- work. In CVPR, pages 1117–1125, 2019. 2, 3, 4, 5, 6
[12] Tsung-Yi Lin, Piotr Dolla ́r, Ross B. Girshick, Kaiming He, Bharath Hariharan, and Serge J. Belongie. Feature pyramid networks for object detection. In CVPR, pages 936–944, 2017. 3
[29] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NeurIPS, pages 5998–6008, 2017. 5
[5] Jiayuan Gu, Han Hu, Liwei Wang, Yichen Wei, and Jifeng Dai. Learning region features for object detection. In ECCV, pages 392–406, 2018. 5
[8] Thomas N. Kipf and Max Welling. Semi-supervised classi- fication with graph convolutional networks. In ICLR, 2017. 5