DocAligner: Annotating Real-world Photographic Document Images by Simply Taking Pictures.
Jiaxin ZhangBangdong ChenHiuyi ChengLianwen JinFengjun GuoKai DingPublished in: CoRR (2023)
Keyphrases
- document images
- real world
- document image analysis
- document processing
- document analysis
- optical character recognition
- page segmentation
- scanned documents
- metadata
- printed documents
- language identification
- historical documents
- page layout
- document image understanding
- word spotting
- mathematical formulas
- word level
- image analysis
- binarization method
- gray scale
- object recognition
- scanned document images