Generative Model Based Noise Robust Training for Unsupervised Domain Adaptation.
Zhongying DengDa LiJunjun HeYi-Zhe SongTao XiangPublished in: CoRR (2023)
Keyphrases
- image noise
- training set
- noisy environments
- imaging artifacts
- salt pepper
- greater robustness
- discriminative training
- training samples
- noisy data
- geometric distortions
- registration errors
- missing data
- noise level
- label noise
- random noise
- training phase
- training algorithm
- training process
- data sets
- partial occlusion
- test set
- computationally efficient
- unsupervised learning
- generative model
- online learning
- data driven
- active learning
- feature selection