all AI news
Feature Alignment by Uncertainty and Self-Training for Source-Free Unsupervised Domain Adaptation. (arXiv:2208.14888v1 [cs.CV])
cs.CV updates on arXiv.org arxiv.org
Most unsupervised domain adaptation (UDA) methods assume that labeled source
images are available during model adaptation. However, this assumption is often
infeasible owing to confidentiality issues or memory constraints on mobile
devices. To address these problems, we propose a simple yet effective
source-free UDA method that uses only a pre-trained source model and unlabeled
target images. Our method captures the aleatoric uncertainty by incorporating
data augmentation and trains the feature generator with two consistency
objectives. The feature generator is encouraged …
alignment arxiv domain adaptation feature free self-training training uncertainty unsupervised