Abnormal event detection via covariance matrix for optical flow based feature.
Tian WangMeina QiaoAichun ZhuYida NiuCe LiHichem SnoussiPublished in: Multim. Tools Appl. (2018)
Keyphrases
- covariance matrix
- optical flow
- abnormal event detection
- covariance matrices
- abnormal events
- principal component analysis
- sample size
- gaussian mixture
- mahalanobis distance
- image sequences
- pseudo inverse
- objective function
- correlation matrix
- positive definite
- geometrical interpretation
- multivariate gaussian
- least squares
- eigenvalues and eigenvectors
- computer vision
- eigendecomposition
- feature set
- cma es
- mahalanobis metric
- estimation error
- daily life
- particle swarm optimization
- upper bound
- search space
- moving objects
- computational complexity
- feature extraction