Feature Alignment by Uncertainty and Self-Training for Source-Free Unsupervised Domain Adaptation

08/31/2022
by   Joonho Lee, et al.
0

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 to learn consistent visual features away from the decision boundaries of the head classifier. Inspired by self-supervised learning, our method promotes inter-space alignment between the prediction space and the feature space while incorporating intra-space consistency within the feature space to reduce the domain gap between the source and target domains. We also consider epistemic uncertainty to boost the model adaptation performance. Extensive experiments on popular UDA benchmarks demonstrate that the performance of our approach is comparable or even superior to vanilla UDA methods without using source images or network modifications.

READ FULL TEXT
research
11/03/2021

Certainty Volume Prediction for Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) deals with the problem of classifyi...
research
11/23/2017

Adversarial Feature Augmentation for Unsupervised Domain Adaptation

Recent works showed that Generative Adversarial Networks (GANs) can be s...
research
02/23/2021

Self-Supervised Noisy Label Learning for Source-Free Unsupervised Domain Adaptation

It is a strong prerequisite to access source data freely in many existin...
research
05/09/2022

Local Prediction Aggregation: A Frustratingly Easy Source-free Domain Adaptation Method

We propose a simple but effective source-free domain adaptation (SFDA) m...
research
03/17/2021

An Efficient Method for the Classification of Croplands in Scarce-Label Regions

Two of the main challenges for cropland classification by satellite time...
research
12/24/2019

Unsupervised Scene Adaptation with Memory Regularization in vivo

We consider the unsupervised scene adaptation problem of learning from b...
research
07/14/2022

Source-Free Domain Adaptation for Real-world Image Dehazing

Deep learning-based source dehazing methods trained on synthetic dataset...

Please sign up or login with your details

Forgot password? Click here to reset