method: Information Extraction from Historical Handwritten Document Images with a Context-aware Neural Mode2018-10-26
Authors: J.Ignacio Toledo, Manuel Carbonell, Alicia Fornés, Josep Lladós
Description: The sequence tagging is performed with a CNN+BLSTM neural network, that accepts a sequence of word images as inputs and produces two independent softmax outputs (person and category) for each word. Those word images tagged as meaningful are then transcribed by generating a sequence of PHOC embeddings that are fed into a BLSTM+CTC network.
Description Paper Source Code
Date | Method | Basic Score | Complete Score | Name | Surname | Location | Occupation | State | Input Type | |||
---|---|---|---|---|---|---|---|---|---|---|---|---|
2018-10-26 | Information Extraction from Historical Handwritten Document Images with a Context-aware Neural Mode | 94.62% | 94.02% | 95.49% | 91.32% | 95.18% | 93.89% | 97.21% | WORD |