HAL: Improved Text-Image Matching by Mitigating Visual Semantic Hubs.
Fangyu LiuRongtian YeXun WangShuaipeng LiPublished in: CoRR (2019)
Keyphrases
- image matching
- semantic content
- semantic information
- image pairs
- arabic text
- matching algorithm
- keypoints
- affine transformation
- semi global
- image retrieval
- wide baseline
- image matching method
- stereo matching
- object recognition
- visual information
- image registration
- object identification
- feature matching
- computer vision
- visual features
- semantic similarity
- wide baseline matching
- invariant features
- semantically related
- matching scheme
- low level features
- image processing
- keywords
- image set
- wordnet
- natural language
- image transformations
- web images
- shape matching
- template matching
- illumination variations
- viewpoint
- scene matching
- image representation
- data mining