PaddlePaddle / Paddle

PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (『飞桨』核心框架,深度学习&机器学习高性能单机、分布式训练和跨平台部署)
http://www.paddlepaddle.org/
Apache License 2.0
22.3k stars 5.62k forks source link

场景文字识别中使用到的模型是CRNN吗? #10517

Closed yeyupiaoling closed 6 years ago

yeyupiaoling commented 6 years ago

在场景文字识别例子https://github.com/PaddlePaddle/models/blob/develop/scene_text_recognition/network_conf.py 中,是使用了CRNN这个模型了,我阅读论文的时候,怎么感觉不太相同。 卷积层:https://github.com/PaddlePaddle/models/blob/develop/scene_text_recognition/network_conf.py#L91-L127 循环层:https://github.com/PaddlePaddle/models/blob/e785f6041379f8399fbb61096bbb0cfa523aa6b5/scene_text_recognition/network_conf.py#L62-L74

那转录层在哪里呢?

qingqing01 commented 6 years ago

@yeyupiaoling 是CRNN模型,使用layer.block_expand将feature map转成序列的。

yeyupiaoling commented 6 years ago

@qingqing01 好的,明白。谢谢您。