Controlling the Output of a Generative Model by Latent Feature Vector Shifting.
Róbert BelanecPeter LackoKristína MalinovskáPublished in: CoRR (2023)
Keyphrases
- generative model
- feature vectors
- probabilistic model
- latent structure
- fisher kernel
- mixture model
- discriminative learning
- em algorithm
- latent variables
- bayesian framework
- latent space
- euclidean distance
- conditional random fields
- discriminative models
- feature space
- posterior probability
- image features
- prior knowledge
- generative process
- semi supervised
- similarity measure
- object categories
- expectation maximization
- latent dirichlet allocation
- hierarchical bayesian model
- decision trees
- image patches
- topic models
- face recognition