Attention-guided adaptation factors for unsupervised facial domain adaptation

In this Letter, the authors introduce attention-guided domain adaptation networks for face recognition under the unsupervised setting. Recently, there has been a dramatic increase in demand for real-world face recognition problems under severe domain shifts. Adversarial learning of the domain adaptation network has shown promising results for this problem. However, this approach has limitations in manually setting the adaptation factor that controls the trade-off between feature discriminability and domain-invariance. To address this critical issue, the authors propose attention-guided domain adaptation networks that can learn the optimal adaptation factor without the manual configurations. The proposed attention-guided module is applied to both category classifier and domain discriminator in a channel-wise manner. By learning the optimal adaptation factor for each channel, the proposed networks successfully align the source domain to the target domain and show state-of-the-art performance on the public datasets.