Sliced Wasserstein Discrepancy in Disentangling Representation and Adaptation Networks for Unsupervised Domain Adaptation
This work addresses domain adaptation for computer vision tasks, but it is incremental as it builds upon an existing method with a modified loss function.
The paper tackled the problem of unsupervised domain adaptation by introducing DRANet-SWD, which uses sliced Wasserstein discrepancy as a style loss to disentangle content and style representations, resulting in enhanced performance on digit classification and driving segmentation benchmarks.
This paper introduces DRANet-SWD as a novel complete pipeline for disentangling content and style representations of images for unsupervised domain adaptation (UDA). The approach builds upon DRANet by incorporating the sliced Wasserstein discrepancy (SWD) as a style loss instead of the traditional Gram matrix loss. The potential advantages of SWD over the Gram matrix loss for capturing style variations in domain adaptation are investigated. Experiments using digit classification datasets and driving scenario segmentation validate the method, demonstrating that DRANet-SWD enhances performance. Results indicate that SWD provides a more robust statistical comparison of feature distributions, leading to better style adaptation. These findings highlight the effectiveness of SWD in refining feature alignment and improving domain adaptation tasks across these benchmarks. Our code can be found here.