Low-dimensional superpixel descriptor for visual correspondence estimation in video.
Songlin DuTakeshi IkenagaPublished in: ISPACS (2017)
Keyphrases
- visual correspondence
- low dimensional
- feature representation
- high dimensional
- multimedia
- video sequences
- content based copy detection
- video data
- input space
- video content
- video streams
- video frames
- mutual information
- data points
- manifold learning
- action detection
- high dimensional data
- real time
- shape descriptors
- multiscale
- feature extraction
- dimensionality reduction
- image retrieval
- object recognition
- euclidean space
- video copy detection
- feature selection
- action recognition
- feature space
- video database
- intrinsic dimensionality
- action classification
- computer vision