method: CNN-LSTM based text detection2017-06-22

Authors: Lulu Xu

Description: The detection results are achieved by a combination of convolutional and recurrent nets. The feature maps are generated by a fully convolutional network based on VGG models. The FCN detects a text line by sliding a window in the last convolutional feature maps and outputs a sequence of a fixed-width text proposals densely. Then the sequential text line proposals are connected by a recurrent neural network.