Adaptive Batch Normalization for practical domain adaptation

Yanghao Li, Naiyan Wang, Jianping Shi, Xiaodi Hou, Jiaying Liu

Abstract: Highlights • A novel domain adaptation technique called Adaptive Batch Normalization (AdaBN). • The effectiveness of AdaBN is validated for both single source and multi-source domain adaptation tasks. • Experiments on the cloud detection for remote sensing images demonstrate the effectiveness of AdaBN in practical use. Abstract Deep neural networks (DNN) have shown unprecedented success in various computer vision applications such as image classification and object detection. However, it is still a common annoyance during the training phase, that one has to prepare at least thousands of labeled images to fine-tune a network to a specific domain. Recent study (Tommasi et al., 2015) shows that a DNN has strong dependency towards the training dataset, and the learned features cannot be easily transferred to a different but relevant task without fine-tuning. In this paper, we propose a simple yet powerful remedy, called Adaptive Batch Normalization (AdaBN) to increase the generalization ability of a DNN. By modulating the statistics from the source domain to the target domain in all Batch Normalization layers across the network, our approach achieves deep adaptation effect for domain adaptation tasks. In contrary to other deep learning domain adaptation methods, our method does not require additional components, and is parameter-free. It archives state-of-the-art performance despite its surprising simplicity. Furthermore, we demonstrate that our method is complementary with other existing methods. Combining AdaBN with existing domain adaptation treatments may further improve model performance. Previous article in issue Next article in issue