Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model.
Richard D. McClanahanPhillip L. De LeonPublished in: Speech Commun. (2015)
Keyphrases
- background model
- background subtraction
- background modeling
- foreground objects
- motion detection
- foreground detection
- moving objects
- gaussian mixture model
- feature vectors
- video sequences
- video surveillance
- surveillance system
- pixel wise
- post processing
- object detection
- computer vision
- appearance model
- graph structure
- moving camera
- segmentation algorithm
- moving target
- three dimensional