Bai, IEEE 2019

Mask TextSpotter

Mask TextSpotter 结构

  • a feature pyramid network (FPN) as backbone.
  • a region proposal network (RPN) for generating text proposals.
  • a Fast R-CNN for bounding boxes regression.
  • a mask branch for text instance segmentation, character segmentation, and text sequence recognition.

数据流动

训练阶段, 由 RPN 产生文本建议(text proposals), 然后文本建议的 RoI 特征喂给 Fast R-CNN, 并且 mask branch 生成精准的文本候选框(candidate boxes), 文本实例分割图(text instance segmentation maps), 字符分割图(character segmentation maps), 和文本序列(text sequence)

posted @ 2019-10-19 09:52  larkii  阅读(117)  评论(0)  编辑  收藏  举报