TextTopicNet - Self-Supervised Learning of Visual Features Through Embedding Images on Semantic Text Spaces.
Yash PatelLluís Gómez i BigordaRaul GomezMarçal RusiñolDimosthenis KaratzasC. V. JawaharPublished in: CoRR (2018)
Keyphrases
- visual features
- web images
- image collections
- image classification
- semantic content
- image retrieval
- low level features and high level
- visual appearance
- semantic gap
- image annotation
- visual and textual features
- low level features
- low level visual features
- bridge the semantic gap
- semantic features
- semantic concepts
- low level
- visual similarity
- visual content
- visual information
- image data
- image database
- image features
- image search
- visual data
- visually similar
- global features
- cbir systems
- labeled images
- keywords
- semantic information
- visual patterns
- active learning
- text queries
- textual descriptions
- image understanding
- high level
- input image
- automatic image annotation
- visual concepts
- bag of features
- multi modal
- image representation
- textual information
- visual attributes
- multimedia
- object recognition
- video retrieval
- image content