Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation
Authors:
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 37 references indexed in Scilit:
- Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-training2018
- Implicit 3D Orientation Learning for 6D Object Detection from RGB Images2018
- Domain Transfer Through Deep Activation Matching2018
- Open Set Domain Adaptation by Backpropagation2018
- Wasserstein Divergence for GANs2018
- DCAN: Dual Channel-Wise Alignment Networks for Unsupervised Scene Adaptation2018
- Sliced and Radon Wasserstein Barycenters of MeasuresJournal of Mathematical Imaging and Vision, 2014
- Microsoft COCO: Common Objects in Context2014
- Evaluation of Traffic Sign Recognition Methods Trained on Synthetically Generated Data2013
- Wasserstein Barycenter and Its Application to Texture Mixing2012