Estimating Viewed Image Categories from fMRI Activity via Multi-view Bayesian Generative Model.
Yusuke AkamatsuRyosuke HarakawaTakahiro OgawaMiki HaseyamaPublished in: GCCE (2019)
Keyphrases
- multi view
- generative model
- bayesian framework
- semi supervised
- single view
- multi views
- depth map
- multiple views
- object categories
- input image
- d objects
- image features
- markov chain monte carlo
- three dimensional
- probabilistic model
- em algorithm
- dynamic scenes
- image classification
- maximum likelihood
- multiscale
- prior knowledge
- topic models
- image regions
- latent dirichlet allocation
- active appearance models
- image segmentation
- image retrieval
- lighting conditions
- similarity measure
- image representation
- feature space
- supervised learning
- pairwise
- posterior distribution
- co training
- learning process
- pixel level
- visual hull
- light field
- high dimensional