Representing Videos Using Mid-level Discriminative Patches.
Arpit JainAbhinav GuptaMikel RodriguezLarry S. DavisPublished in: CVPR (2013)
Keyphrases
- mid level
- low level
- mid level features
- low level features
- high level
- action recognition
- event detection
- visual information
- superpixels
- visual cues
- levels of abstraction
- perceptual grouping
- visual concepts
- human actions
- natural images
- image patches
- video content
- key frames
- video sequences
- video frames
- video analysis
- video data
- visual data
- semantic concepts
- higher level
- multiscale
- relevance feedback
- eye movements
- feature vectors
- segmentation algorithm
- visual features
- learning algorithm