Domain Invariant Regularization By Disentangling content and style features for Visual Domain Generalization

In this paper, taking the advantage of multiple source domains, we propose a novel approach for visual Domain Generalization (DG). The three key ideas underlying our formulation are (1) leveraging disentangled representations of the images to define different factors of variations, (2) generating perturbed images by changing such factors composing the representations of the images, (3) enforcing the learner (classifier) to be invariant to such changes in the images. We demonstrate the effectiveness of our approach on several widely used datasets for the domain generalization problem, on all of which we achieve competitive results with state-of-the-art models.