Image classification based on log-Euclidean Fisher Vectors for covariance matrix descriptors.
Sara AkodadLionel BombrunCharles YaacoubYannick BerthoumieuChristian GermainPublished in: IPTA (2018)
Keyphrases
- fisher vectors
- image classification
- covariance matrix
- log euclidean
- covariance matrices
- affine invariant
- riemannian metric
- bag of words
- vector space
- visual features
- sample size
- principal component analysis
- image representation
- image features
- feature extraction
- bag of features
- tensor field
- riemannian manifolds
- lie group
- visual words
- svm classifier
- euclidean space
- gaussian mixture
- scene classification
- multi class
- diffusion tensor
- distance metric
- objective function
- dimensionality reduction
- multiscale