Bi-Directional Generation for Unsupervised Domain Adaptation

TL;DR

This paper introduces a bi-directional generation model for unsupervised domain adaptation that balances domain gap reduction with data structure preservation, leading to improved performance on visual benchmarks.

Contribution

It proposes a novel bi-directional generation approach with consistent classifiers and cross-domain alignment, enhancing domain adaptation without destroying data structure.

Findings

Outperforms state-of-the-art on visual benchmarks

Effectively balances domain gap mitigation and data structure preservation

Uses cross-domain generators and consistent classifiers

Abstract

Unsupervised domain adaptation facilitates the unlabeled target domain relying on well-established source domain information. The conventional methods forcefully reducing the domain discrepancy in the latent space will result in the destruction of intrinsic data structure. To balance the mitigation of domain gap and the preservation of the inherent structure, we propose a Bi-Directional Generation domain adaptation model with consistent classifiers interpolating two intermediate domains to bridge source and target domains. Specifically, two cross-domain generators are employed to synthesize one domain conditioned on the other. The performance of our proposed method can be further enhanced by the consistent classifiers and the cross-domain alignment constraints. We also design two classifiers which are jointly optimized to maximize the consistency on target sample prediction. Extensive experiments verify that our proposed model outperforms the state-of-the-art on standard cross domain visual benchmarks.