Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization.
Daiqing LiJunlin YangKarsten KreisAntonio TorralbaSanja FidlerPublished in: CoRR (2021)
Keyphrases
- semi supervised learning
- discriminative models
- generative model
- semantic segmentation
- semi supervised
- conditional random fields
- object categories
- unlabeled data
- labeled data
- weakly supervised
- transfer learning
- superpixels
- supervised learning
- co training
- active learning
- machine learning
- training data
- mixture model
- probabilistic model
- unsupervised learning
- expectation maximization
- object class
- em algorithm
- topic models
- hidden markov models
- prior knowledge
- pairwise
- text classification
- markov random field
- higher order
- object classes
- bounding box
- similarity measure
- feature selection