Expressing Objects Just Like Words: Recurrent Visual Embedding for Image-Text Matching.
Tianlang ChenJiebo LuoPublished in: AAAI (2020)
Keyphrases
- semantic labels
- visual appearance
- keypoints
- auto annotation
- web images
- image data
- image regions
- image matching
- image features
- image content
- template matching
- matching process
- spatial relations
- visual data
- single image
- low level
- image pixels
- visual features
- visually similar
- visual perception
- image classification
- visual attributes
- image retrieval
- feature points
- input image
- image set
- image collections
- spatial layout
- low level image features
- multiscale
- image segmentation
- keywords
- d objects
- multiple objects
- image annotation
- spatial arrangement
- scale invariant features
- spatial information
- object models
- complex scenes
- spatial relationships
- text documents
- matching algorithm
- complex background
- visual patterns
- test images
- pixel values
- visual scene
- bounding box
- text mining
- text queries
- false matches
- semantic information
- handwritten words
- spatial configuration
- visual input
- image representation
- rigid transformation
- visual similarity
- feature descriptors