VSR++: Improving Visual Semantic Reasoning for Fine-Grained Image-Text Matching.
Hui YuanYan HuangDongbo ZhangZerui ChenWenlong ChengLiang WangPublished in: ICPR (2020)
Keyphrases
- fine grained
- coarse grained
- image matching
- input image
- semantic labels
- low level
- feature points
- image features
- image classification
- web images
- image content
- auto annotation
- semantic content
- matching process
- keypoints
- image set
- image representation
- visual features
- visual concepts
- semantic space
- image segmentation
- textual descriptions
- access control
- image retrieval
- knowledge base
- information retrieval
- high level
- spatial relations
- semantic information
- keywords
- image collections
- tightly coupled
- low level features