Light-weight Calibrator: a Separable Component for Unsupervised Domain Adaptation
This addresses domain shift problems in computer vision, offering a lightweight solution that preserves source performance while adapting to target domains, though it appears incremental by building on existing GAN-based methods.
The paper tackles unsupervised domain adaptation by introducing a separable data calibrator that helps a fixed source classifier maintain performance in the target domain without retraining, achieving state-of-the-art results on digits datasets and driving scene semantic segmentation.
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 empirically reveals that certain intriguing hints, which can be mitigated by adversarial attack to domain discriminators, are one of the sources for performance degradation under the domain shift.