Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation

Abstract

Domain adaptation (DA) aims to alleviate the domain shift between source domain and target domain. Most DA methods require access to the source data, but often that is not possible (e.g. due to data privacy or intellectual property). In this paper, we address the challenging source-free domain adaptation (SFDA) problem, where the source pretrained model is adapted to the target domain in the absence of source data. Our method is based on the observation that target data, which might no longer align with the source domain classifier, still forms clear clusters. We capture this intrinsic structure by defining local affinity of the target data, and encourage label consistency among data with high local affinity. We observe that higher affinity should be assigned to reciprocal neighbors, and propose a self regularization loss to decrease the negative impact of noisy neighbors. Furthermore, to aggregate information with more context, we consider expanded neighborhoods with small affinity values. In the experimental results we verify that the inherent structure of the target features is an important source of information for domain adaptation. We demonstrate that this local structure can be efficiently captured by considering the local neighbors, the reciprocal neighbors, and the expanded neighborhood. Finally, we achieve state-of-the-art performance on several 2D image and 3D point cloud recognition datasets. Code is available in https://github.com/Albert0147/SFDA_neighbors.

NeurIPS 2021. Fix four number errors in Tab.5 (first two tables, row 3 and 4) and corresponding text

References in corpus (7)

• Deep Domain Confusion: Maximizing for Domain Invariance
• VisDA: The Visual Domain Adaptation Challenge
• Model Adaptation: Unsupervised Domain Adaptation without Source Data
• Bridging Theory and Algorithm for Domain Adaptation
• Information-Theoretical Learning of Discriminative Clusters for Unsupervised Domain Adaptation
• PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation
• Dual Mixup Regularized Learning for Adversarial Domain Adaptation

Cited by in corpus (4)

• Uncertainty-Induced Transferability Representation for Source-Free Unsupervised Domain Adaptation
• Collaborate to Adapt: Source-Free Graph Domain Adaptation via Bi-directional Adaptation
• Source Free Unsupervised Graph Domain Adaptation
• GIPSO: Geometrically Informed Propagation for Online Adaptation in 3D LiDAR Segmentation