TextManiA: Enriching Visual Feature by Text-driven Manifold Augmentation.
Moon Ye-BinJisoo KimHongyeob KimKilho SonTae-Hyun OhPublished in: CoRR (2023)
Keyphrases
- visual features
- keywords
- web images
- semantic content
- visual information
- low level features
- image classification
- content based video retrieval
- image search
- automatic image annotation
- visual content
- image retrieval
- semantic information
- image annotation
- semantic concepts
- semantic features
- information retrieval
- low level
- text documents
- global features
- semantic gap
- visual similarity
- multimedia data
- image content
- high dimensional
- image features