This repository lets you train neural networks models for performing end-to-end full-page handwriting recognition using the Apache MXNet deep learning frameworks on the IAM Dataset.
Apache License 2.0
481
stars
189
forks
source link
Can this OCR be used for japanese line prediction too? #34
It may be possible to train the architecture for hiragana or katakana if you have a labelled dataset. However, the current CNN-biLSTM is not optimised for kanji recognition (please see https://ieeexplore.ieee.org/document/8977956).
It may be possible to train the architecture for hiragana or katakana if you have a labelled dataset. However, the current CNN-biLSTM is not optimised for kanji recognition (please see https://ieeexplore.ieee.org/document/8977956).