Light-weight Calibrator: A Separable Component for Unsupervised Domain Adaptation

Shaokai Ye, Kailu Wu, Mu Zhou, Yunfei Yang, Sia Huat Tan, Kaidi Xu, Jiebo Song, Chenglong Bao, Kaisheng Ma; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 13736-13745

Abstract


Existing domain adaptation methods aim at learning features that can be generalized among domains. These methods commonly require to update source classifier to adapt to the target domain and do not properly handle the trade-off between the source domain and the target domain. In this work, instead of training a classifier to adapt to the target domain, we use a separable component called data calibrator to help the fixed source classifier recover discrimination power in the target domain, while preserving the source domain's performance. When the difference between two domains is small, the source classifier's representation is sufficient to perform well in the target domain and outperforms GAN-based methods in digits. Otherwise, the proposed method can leverage synthetic images generated by GANs to boost performance and achieve state-of-the-art performance in digits datasets and driving scene semantic segmentation. Our method also empirically suggests the potential connection between domain adaptation and adversarial attacks. Code release is available at https://github.com/yeshaokai/ Calibrator-Domain-Adaptation

Related Material


[pdf] [supp] [arXiv] [video]
[bibtex]
@InProceedings{Ye_2020_CVPR,
author = {Ye, Shaokai and Wu, Kailu and Zhou, Mu and Yang, Yunfei and Tan, Sia Huat and Xu, Kaidi and Song, Jiebo and Bao, Chenglong and Ma, Kaisheng},
title = {Light-weight Calibrator: A Separable Component for Unsupervised Domain Adaptation},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}