BriefGPT.xyz
Jun, 2015
在深度卷积序列中阅读场景文本
Reading Scene Text in Deep Convolutional Sequences
HTML
PDF
Pan He, Weilin Huang, Yu Qiao, Chen Change Loy, Xiaoou Tang
TL;DR
使用深度文本递归网络(DTRN)进行场景文本识别,包含顺序标记问题,卷积神经网络,长短时记忆 lstm 模型,具有高度的识别准确率,对于高度不确定的单词具有可靠性等优点。
Abstract
We develop a
deep-text recurrent network
(DTRN) that regards scene text reading as a
sequence labelling
problem. We leverage recent advances of deep
→