G-LBM: Generative Low-Dimensional Background Model Estimation from Video Sequences.
Behnaz RezaeiAmirreza FarnooshSarah OstadabbasPublished in: ECCV (12) (2020)
Keyphrases
- background model
- video sequences
- foreground objects
- low dimensional
- background subtraction
- moving objects
- dynamic background
- background modeling
- high dimensional
- image sequences
- motion detection
- dynamic scenes
- moving object detection
- moving camera
- dimensionality reduction
- object segmentation
- object tracking
- video surveillance
- subspace learning
- motion estimation
- frame rate
- pixel level
- video data
- template matching
- video frames
- data points
- principal component analysis
- manifold learning
- feature space
- surveillance system
- space time
- segmentation algorithm
- generative model
- motion segmentation
- object detection
- multi class
- video camera
- high dimensional data
- detection algorithm
- active contours
- unsupervised learning
- input image
- three dimensional