FixBi: Bridging Domain Spaces for Unsupervised Domain Adaptation

CVPR 2021  ·  Jaemin Na, Heechul Jung, Hyung Jin Chang, Wonjun Hwang ·

Unsupervised domain adaptation (UDA) methods for learning domain invariant representations have achieved remarkable progress. However, most of the studies were based on direct adaptation from the source domain to the target domain and have suffered from large domain discrepancies... In this paper, we propose a UDA method that effectively handles such large domain discrepancies. We introduce a fixed ratio-based mixup to augment multiple intermediate domains between the source and target domain. From the augmented-domains, we train the source-dominant model and the target-dominant model that have complementary characteristics. Using our confidence-based learning methodologies, e.g., bidirectional matching with high-confidence predictions and self-penalization using low-confidence predictions, the models can learn from each other or from its own results. Through our proposed methods, the models gradually transfer domain knowledge from the source to the target domain. Extensive experiments demonstrate the superiority of our proposed method on three public benchmarks: Office-31, Office-Home, and VisDA-2017. read more

PDF Abstract CVPR 2021 PDF CVPR 2021 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Domain Adaptation Office-31 FixBi Average Accuracy 91.4 # 1
Domain Adaptation Office-Home FixBi Accuracy 72.7 # 1
Domain Adaptation VisDA2017 FixBi Accuracy 87.2 # 1