Zero-Shot Image-to-Text Generation for Visual-Semantic Arithmetic.
Yoad TewelYoav ShalevIdan SchwartzLior WolfPublished in: CoRR (2021)
Keyphrases
- input image
- text generation
- image data
- visual perception
- low level
- single image
- visually similar
- image features
- image classification
- visual appearance
- image retrieval
- image content
- visual concepts
- semantic space
- multiscale
- visual cues
- low level visual features
- image regions
- image segmentation
- test images
- auto annotation
- visual information
- visual similarity
- visual attributes
- natural language generation
- spatial relations
- image representation
- visual features
- natural language
- high level
- web images
- image collections
- domain specific
- expert systems
- scene categorization
- high level semantics
- similarity measure