Looking at words and points with attention: a benchmark for text-to-shape coherence.
Andrea AmaduzziGiuseppe LisantiSamuele SaltiLuigi Di StefanoPublished in: ICCV (Workshops) (2023)
Keyphrases
- text documents
- keywords
- english words
- medial axis
- related words
- text recognition
- short text
- sample points
- linguistic information
- text databases
- proper nouns
- text corpus
- chinese text
- word pairs
- text mining
- syntactic categories
- shape descriptors
- text corpora
- world knowledge
- linguistic analysis
- chinese texts
- shape features
- text representation
- textual features
- hand drawn
- information retrieval
- salient points
- natural language text
- feature points
- shape model
- arabic text
- lexical features
- noun phrases
- polygonal approximation
- text retrieval
- punctuation marks
- closed curves
- data points
- text categorization
- n gram
- shape analysis
- multiword
- boundary points
- shape matching
- arbitrary shape
- word recognition
- word level
- document analysis
- natural language processing
- language model
- arabic language
- text classification
- object shapes
- training corpus
- shape representation
- semantic information
- visual attention
- word sense disambiguation
- landmark points