Consensus-Aware Visual-Semantic Embedding for Image-Text Matching.
Haoran WangYing ZhangZhong JiYanwei PangLin MaPublished in: ECCV (24) (2020)
Keyphrases
- web images
- input image
- image matching
- template matching
- semantic content
- image content
- matching process
- keypoints
- semantic labels
- visual concepts
- low level
- feature points
- multiscale
- visual perception
- image data
- high level
- auto annotation
- scene matching
- image classification
- image retrieval
- image set
- semantic space
- feature matching
- image features
- string matching
- image representation
- visual features
- semantic matching
- semantic information
- visually similar
- matching scheme
- high resolution
- low level visual features
- image segmentation
- textual descriptions
- natural language
- arabic text
- visual data
- edge detection
- image annotation
- semantically related
- semantic gap
- visual information
- visual attributes
- cover image
- keywords
- semantic context
- similarity measure
- matching algorithm
- image regions