Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition.
Yue HeChen ChenJing ZhangJuhua LiuFengxiang HeChaoyue WangBo DuPublished in: AAAI (2022)
Keyphrases
- scene text recognition
- visual representations
- proof theory
- reasoning tasks
- open world
- visual information
- computational properties
- object recognition
- formal semantics
- default reasoning
- multimedia
- graphical representation
- logic programming
- low level
- visual perception
- visual features
- automated reasoning
- default rules
- proof theoretic
- semantic information
- logical rules
- high level
- logical framework
- visual representation
- visual content
- highly expressive
- defeasible reasoning
- high level semantics
- reasoning mechanisms
- derivation rules
- fuzzy description logics
- metadata
- uml class diagrams
- reasoning mechanism
- reasoning problems
- knowledge representation
- knowledge representation and reasoning
- reasoning systems
- logic programs
- uncertain information