Uncertainty-aware consistency regularization for cross-domain semantic segmentation.
Qianyu ZhouZhengyang FengQiqi GuGuangliang ChengXuequan LuJianping ShiLizhuang MaPublished in: Comput. Vis. Image Underst. (2022)
Keyphrases
- cross domain
- semantic segmentation
- superpixels
- conditional random fields
- scene classification
- weakly supervised
- text categorization
- knowledge transfer
- transfer learning
- object categories
- object classes
- image understanding
- e government
- target domain
- feature selection
- segmentation method
- learning algorithm
- image classification
- data points
- probabilistic model
- object recognition
- image segmentation