Semantic Segmentation With Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization.
Daiqing LiJunlin YangKarsten KreisAntonio TorralbaSanja FidlerPublished in: CVPR (2021)
Keyphrases
- semi supervised learning
- discriminative models
- generative model
- semantic segmentation
- semi supervised
- conditional random fields
- object categories
- unlabeled data
- weakly supervised
- labeled data
- transfer learning
- unsupervised learning
- probabilistic model
- superpixels
- machine learning
- supervised learning
- training data
- mixture model
- pairwise
- topic models
- prior knowledge
- co training
- information retrieval
- active learning
- object classes
- higher order
- hidden markov models
- text classification
- em algorithm
- object recognition
- scene classification
- object class
- natural images
- graphical models
- image classification
- data sets