Automatic Recovery of Corrupted Font Encoding in PDF Documents Using CNN-based Symbol Recognition with Language Model.
Mark VolAndrey KrutkoNicolas StefanovitchDenis PostanogovPublished in: DAS (2018)
Keyphrases
- language model
- symbol recognition
- language modeling
- n gram
- probabilistic model
- document retrieval
- context sensitive
- speech recognition
- retrieval model
- information retrieval
- query expansion
- pdf documents
- test collection
- mixture model
- query terms
- ad hoc information retrieval
- feature selection
- vector space model
- pseudo relevance feedback
- document image analysis
- character recognition
- translation model
- smoothing methods
- user interface
- word clouds