LSTM for Image Annotation with Relative Visual Importance.
Geng YanYang WangZicheng LiaoPublished in: BMVC (2016)
Keyphrases
- image annotation
- visual features
- web images
- automatic image annotation
- image search
- visual information
- multi modal
- visual similarity
- image annotation and retrieval
- image understanding
- multi label
- automatic annotation
- visual concepts
- web image search
- image retrieval
- image content
- low level features
- image classification
- high level semantics
- annotated images
- semantic gap
- visual and textual information
- image collections
- low level
- visual content
- low level visual features
- multi label image annotation
- semantic concepts
- high level
- computer vision
- relevance model
- relevance feedback
- keywords
- training data
- search engine