Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation
This work addresses domain shift problems in computer vision, offering a method for adapting models across different data distributions, but it appears incremental as it builds on existing concepts of distribution alignment and Wasserstein metrics.
The paper tackled unsupervised domain adaptation by connecting feature distribution alignment with the Wasserstein metric, proposing sliced Wasserstein discrepancy (SWD) to measure dissimilarity between classifier outputs and guide alignment, achieving validation across tasks like digit recognition and image classification.
In this work, we connect two distinct concepts for unsupervised domain adaptation: feature distribution alignment between domains by utilizing the task-specific decision boundary and the Wasserstein metric. Our proposed sliced Wasserstein discrepancy (SWD) is designed to capture the natural notion of dissimilarity between the outputs of task-specific classifiers. It provides a geometrically meaningful guidance to detect target samples that are far from the support of the source and enables efficient distribution alignment in an end-to-end trainable fashion. In the experiments, we validate the effectiveness and genericness of our method on digit and sign recognition, image classification, semantic segmentation, and object detection.