Contribution of recurrent connectionist language models in improving LSTM-based Arabic text recognition in videos
作者：
Highlights：
• Different recurrent connectionist language models to improve LSTM-based Arabic text recognition in videos.
• Efficient joint decoding paradigm using language model and LSTM responses.
• Additional decoding hyper-parameters, extensively evaluated, that improve recognition results and optimize running time.
• Significant recognition improvement by integrating connectionist language models that outperform n-grams contribution.
• Final Arabic OCR system that significantly outperforms commercial OCR engine.
摘要
Highlights•Different recurrent connectionist language models to improve LSTM-based Arabic text recognition in videos.•Efficient joint decoding paradigm using language model and LSTM responses.•Additional decoding hyper-parameters, extensively evaluated, that improve recognition results and optimize running time.•Significant recognition improvement by integrating connectionist language models that outperform n-grams contribution.•Final Arabic OCR system that significantly outperforms commercial OCR engine.
论文关键词：Connectionist language model,Arabic video OCR,Decoding,RNN,LSTM,Convolutional Neural Network
论文评审过程：Received 11 March 2016, Revised 12 November 2016, Accepted 15 November 2016, Available online 16 November 2016, Version of Record 28 November 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.11.011
