Variational Autoencoders Pursue PCA Directions (by Accident).
Michal RolinekDominik ZietlowGeorg MartiusPublished in: CVPR (2019)
Keyphrases
- principal component analysis
- denoising
- principal components analysis
- principal components
- dimensionality reduction
- principle component analysis
- independent component analysis
- dimension reduction
- face recognition
- feature extraction
- optical flow
- covariance matrix
- image segmentation
- appearance based object recognition
- variational methods
- k means
- kernel pca
- methods in computer vision
- discriminant analysis
- low dimensional
- face images
- approximate inference
- lower dimensional
- subspace methods
- free energy
- high dimensional
- computer vision