Joint Geometrical and Statistical Alignment for Visual Domain Adaptation

TL;DR

This paper introduces JGSA, a novel unsupervised domain adaptation method that simultaneously reduces geometrical and statistical domain shifts, improving cross-domain visual recognition performance.

Contribution

The paper proposes a unified framework with coupled projections for joint geometrical and statistical alignment, solved efficiently in closed form.

Findings

Outperforms state-of-the-art methods on multiple datasets

Effectively reduces domain shift in visual recognition tasks

Validated on synthetic and real-world datasets

Abstract

This paper presents a novel unsupervised domain adaptation method for cross-domain visual recognition. We propose a unified framework that reduces the shift between domains both statistically and geometrically, referred to as Joint Geometrical and Statistical Alignment (JGSA). Specifically, we learn two coupled projections that project the source domain and target domain data into low dimensional subspaces where the geometrical shift and distribution shift are reduced simultaneously. The objective function can be solved efficiently in a closed form. Extensive experiments have verified that the proposed method significantly outperforms several state-of-the-art domain adaptation methods on a synthetic dataset and three different real world cross-domain visual recognition tasks.