Multi-Level Visual-Semantic Alignments with Relation-Wise Dual Attention Network for Image and Text Matching.
Zhibin HuYongsheng LuoJiong LinYan YanJian ChenPublished in: IJCAI (2019)
Keyphrases
- web images
- input image
- template matching
- semantic space
- semantic content
- image matching
- auto annotation
- image content
- visually similar
- matching process
- image features
- image classification
- keypoints
- image collections
- visual concepts
- image segmentation
- image retrieval
- feature points
- low level
- single image
- image representation
- visual appearance
- visual data
- scene matching
- image data
- multiscale
- textual descriptions
- keywords
- semantic network
- visual attributes
- low level visual features
- semantic labels
- high resolution
- semantic information
- matching scheme
- semantic matching
- feature matching
- image set
- visual features
- matching algorithm
- arabic text
- visual cues
- semantic representations
- high level
- visual perception
- semantic gap
- saliency map
- text mining
- visual attention
- semantically related
- semantic concepts
- image search
- low level features
- dependency relations
- co occurrence
- image regions
- pattern matching