Adaptation Across Extreme Variations using Unlabeled Domain Bridges
This work tackles domain adaptation for scenarios where source and target domains are highly dissimilar, which is a common challenge in real-world applications like computer vision.
The paper addresses unsupervised domain adaptation with large domain discrepancies by introducing unlabeled bridging domains to decompose the discrepancy into smaller, more manageable parts, and validates the method on tasks like object recognition and semantic segmentation.
We tackle an unsupervised domain adaptation problem for which the domain discrepancy between labeled source and unlabeled target domains is large, due to many factors of inter and intra-domain variation. While deep domain adaptation methods have been realized by reducing the domain discrepancy, these are difficult to apply when domains are significantly unalike. In this work, we propose to decompose domain discrepancy into multiple but smaller, and thus easier to minimize, discrepancies by introducing unlabeled bridging domains that connect the source and target domains. We realize our proposal through an extension of the domain adversarial neural network with multiple discriminators, each of which accounts for reducing discrepancies between unlabeled (bridge, target) domains and a mix of all precedent domains including source. We validate the effectiveness of our method on several adaptation tasks including object recognition and semantic segmentation.