Unsupervised Cross-domain Image Classification by Distance Metric Guided Feature Alignment
This work addresses domain adaptation for medical imaging, specifically cross-device fetal ultrasound classification, but is incremental as it builds on existing feature alignment techniques.
The paper tackles the problem of domain shift in unsupervised domain adaptation for image classification by proposing a method that learns discriminative and domain-invariant features without adversarial training, and it outperforms state-of-the-art methods on fetal ultrasound datasets.
Learning deep neural networks that are generalizable across different domains remains a challenge due to the problem of domain shift. Unsupervised domain adaptation is a promising avenue which transfers knowledge from a source domain to a target domain without using any labels in the target domain. Contemporary techniques focus on extracting domain-invariant features using domain adversarial training. However, these techniques neglect to learn discriminative class boundaries in the latent representation space on a target domain and yield limited adaptation performance. To address this problem, we propose distance metric guided feature alignment (MetFA) to extract discriminative as well as domain-invariant features on both source and target domains. The proposed MetFA method explicitly and directly learns the latent representation without using domain adversarial training. Our model integrates class distribution alignment to transfer semantic knowledge from a source domain to a target domain. We evaluate the proposed method on fetal ultrasound datasets for cross-device image classification. Experimental results demonstrate that the proposed method outperforms the state-of-the-art and enables model generalization.