Light-weight Calibrator: a Separable Component for Unsupervised Domain Adaptation
| Autor: | Shaokai Ye, Kailu Wu, Mu Zhou, Jiebo Song, Yunfei Yang, Chenglong Bao, Kaidi Xu, Sia Huat Tan, Kaisheng Ma |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Domain adaptation Computer Science - Machine Learning Artificial neural network business.industry Computer science Computer Vision and Pattern Recognition (cs.CV) 05 social sciences Feature extraction Image and Video Processing (eess.IV) Computer Science - Computer Vision and Pattern Recognition Pattern recognition 010501 environmental sciences Electrical Engineering and Systems Science - Image and Video Processing 01 natural sciences Data modeling Separable space Machine Learning (cs.LG) 0502 economics and business FOS: Electrical engineering electronic engineering information engineering Leverage (statistics) Artificial intelligence 050207 economics business 0105 earth and related environmental sciences |
| Zdroj: | CVPR |
| Popis: | 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. Accepted by CVPR2020 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|