Source-Free Domain Adaptation for Medical Image Segmentation via Prototype-Anchored Feature Alignment and Contrastive Learning.
Qinji YuNan XiJunsong YuanZiyu ZhouKang DangXiaowei DingPublished in: CoRR (2023)
Keyphrases
- domain adaptation
- medical image segmentation
- multiple sources
- learning algorithm
- learning process
- covariate shift
- medical images
- prior knowledge
- semi supervised learning
- transfer learning
- target domain
- supervised learning
- semi supervised
- unsupervised learning
- active learning
- pairwise
- source domain
- manifold alignment
- machine learning