SelfDocSeg: A Self-Supervised vision-based Approach towards Document Segmentation.
Subhajit MaitySanket BiswasSiladittya MannaAyan BanerjeeJosep LladósSaumik BhattacharyaUmapada PalPublished in: CoRR (2023)
Keyphrases
- medical imaging
- medical images
- computer vision
- image segmentation
- information retrieval systems
- document images
- information retrieval
- real time
- keywords
- vision system
- document collections
- relevant documents
- image regions
- document classification
- shape prior
- retrieval systems
- segmentation algorithm
- level set
- energy function
- web documents
- region growing
- multiple objects
- vector space model
- segmentation accuracy
- low level vision
- cf loadingtexthtml