Mean Box Pooling: A Rich Image Representation and Output Embedding for the Visual Madlibs Task.
Ashkan MokarianMateusz MalinowskiMario FritzPublished in: CoRR (2016)
Keyphrases
- image representation
- image classification
- multiscale
- scene categorization
- bag of words
- object recognition
- visual vocabulary
- spatial pyramid matching
- visual words
- image retrieval
- visual features
- quadtree
- image content
- bag of visual words
- bag of features
- visual information
- low level features
- image features
- sparse representation
- representation scheme
- scene classification
- scene recognition
- low level
- feature space
- scene categories
- feature representations
- spatial pyramid
- spatial layout
- mid level
- high level
- feature extraction
- pattern representation
- receptive fields
- sparse coding
- natural images
- high dimensional