Mismatch Quest: Visual and Textual Feedback for Image-Text Misalignment.
Brian GordonYonatan BittonYonatan ShafirRoopal GargXi ChenDani LischinskiDaniel Cohen-OrIdan SzpektorPublished in: CoRR (2023)
Keyphrases
- web images
- image data
- input image
- low level
- image features
- image analysis
- single image
- image segmentation
- textual descriptions
- multiscale
- visual perception
- text information
- edge detection
- image classification
- pixel values
- visual data
- visual attributes
- textual information
- visual appearance
- textual data
- textual query
- auto annotation
- image representation
- image regions
- relevance feedback
- high resolution
- test images
- image collections
- information retrieval
- visual similarity
- visual concepts
- visual representations
- visually similar
- visual information
- scanned documents
- visual features
- textual features
- similarity measure
- visual input
- plain text
- textual and visual information
- visual and textual features
- visual cues
- text documents
- spatial information
- segmentation method
- text mining
- keywords
- high level
- image processing
- metadata