Authors: Qianyun Chu, Peiyao Zhang, Xixi Xu, Guanxian Chen, Jingye Chen, Jianqi Ma, Bin Li, Xiangyang Xue
Description: We use a multi-object rectified attention network (MORAN) for general scene text recognition.The MORAN consists of a multi-object rectification network (MORN) to rectify images and an attention-based sequence recognition network (ASRN) to read the text. And furthermore, we do some changes. At the same time, we also use a new method called GRCNN for general scene text recognition. The most important module is GRCL, it helps to improve the accuracy in great degree.
Here，we made some model fusion，more exactly，the fusion of MORAN and GRCNN.