Multi-modal visual concept classification of images via Markov random walk over tags.
Motoaki KawanabeAlexander BinderChristina MüllerWojciech WojcikiewiczPublished in: WACV (2011)
Keyphrases
- multi modal
- visual concepts
- image annotation
- semantic concepts
- visual similarity
- image collections
- semantic gap
- image classification
- single modality
- automatic image annotation
- web images
- image data
- image database
- visual features
- markov random walk
- image search
- high dimensional
- input image
- multiple modalities
- feature extraction
- visual content
- image features
- visual information
- learning tasks
- image content
- object categories
- bag of words
- image registration
- feature selection
- object recognition
- machine learning
- object detection
- metadata
- higher level
- multiscale
- reinforcement learning
- random walk
- visual data
- probabilistic model