Classifying spatiotemporal object trajectories using unsupervised learning in the coefficient feature space.
Andrew NaftelShehzad KhalidPublished in: Multim. Syst. (2006)
Keyphrases
- object trajectories
- unsupervised learning
- feature space
- unsupervised manner
- spatio temporal
- dimensionality reduction
- video surveillance
- multiple cameras
- motion model
- feature selection
- surveillance videos
- supervised learning
- high dimensional
- static images
- mean shift
- activity analysis
- feature vectors
- semi supervised
- image representation
- feature extraction
- motion patterns
- object recognition
- principal component analysis
- text classification
- moving objects
- machine learning
- data points
- computer vision
- trajectory data
- field of view
- event detection
- model selection
- optical flow
- training set
- three dimensional
- visual words
- object tracking
- image processing
- multi view
- object classification
- motion estimation
- image segmentation
- image sequences