DocParser: End-to-end OCR-free Information Extraction from Visually Rich Documents.
Mohamed DhouibGhassen BettaiebAymen ShabouPublished in: CoRR (2023)
Keyphrases
- end to end
- information extraction
- document processing
- text documents
- free text
- web documents
- printed documents
- text localization and recognition
- unstructured documents
- information retrieval
- scanned documents
- document analysis
- text mining
- page layout
- natural language processing
- congestion control
- admission control
- document images
- optical character recognition
- wireless ad hoc networks
- character recognition
- transport layer
- multipath
- ad hoc networks
- high bandwidth
- question answering
- content delivery
- internet protocol
- scanned images
- scalable video
- named entities
- video coding
- application layer
- metadata
- machine learning