Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation From Unlabeled Data.
David S. W. WilliamsDaniele De MartiniMatthew GaddPaul NewmanPublished in: IEEE Trans. Robotics (2024)
Keyphrases
- unlabeled data
- semantic segmentation
- labeled data
- semi supervised learning
- semi supervised
- supervised methods
- weakly supervised
- training data
- supervised learning
- conditional random fields
- active learning
- labeled training data
- text classification
- data points
- superpixels
- training examples
- learning algorithm
- scene classification
- object categories
- text categorization
- class labels
- training set
- co occurrence
- pascal voc
- multiscale
- object classes
- generative model
- unsupervised learning
- pairwise
- machine learning
- topic models
- data sets
- multi view
- input image
- object recognition
- reinforcement learning
- three dimensional