method: NXB OCR2019-06-03

Authors: :Cheng Cheng*(N), Jing Li(X), Qi Qu(B), Haoyuan Wang(N), Qiufeng Wang*(X), YuPeng Cao(X), Jie Zhang(B), Ming Zhang(B), Kaizhu Huang*(X)(EqualContribution)Xuguang Wang(N)

Description: The recognition part is based on MORAN-v2 [1] and CRNN [2], the former consisting of a rectification network and an Attention-based recognition network, while the latter adopting a CTC-based algorithm. It is trained on synthetic data and fine-tuned on MLT 2019 training set with both variable length.

P.S.Affiliation of Authors
(N:Institute of Nanotechnology and Nano-Bionics, Chinese Academy of Sciences ;
X:Xi’an Jiaotong-liverpool University ;
B:Beijing Babel Tenchnology Co., Ltd.)

[1] C. Luo, L. Jin, and Z. Sun, "MORAN: A Multi-Object Rectified Attention Network for scene text recognition," Pattern Recognition, vol. 90, pp. 109-118, 2019. [2] B. Shi, X. Bai, and C. Yao, "An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition," IEEE transactions on pattern analysis and machine intelligence, vol. 39, no. 11, pp. 2298-2304, 2016.