Textual restoration of occluded Tibetan document pages based on side-enhanced U-Net.
Siqi LiuLibiao JinFang MiaoPublished in: J. Electronic Imaging (2020)
Keyphrases
- keywords
- html pages
- web documents
- website
- web pages
- textual content
- html documents
- search engine
- www pages
- textual information
- document classification
- image restoration
- document structure
- information retrieval
- textual features
- text content
- text documents
- structured documents
- information retrieval systems
- retrieval systems
- semi structured
- natural language
- wikipedia pages
- document analysis
- textual data
- document images
- document representation
- document collections
- textual contents
- printed text
- page layout
- image processing
- multimedia
- web content
- web users
- document clustering
- user queries
- focused crawling
- web search engines
- metadata
- co occurrence
- visual features
- text mining