OSSA: Unsupervised One-Shot Style Adaptation

TL;DR

OSSA introduces an unsupervised one-shot style adaptation method for object detection that leverages a single unlabeled target image to improve performance across various domain shifts, outperforming existing methods.

Contribution

The paper proposes OSSA, a novel unsupervised domain adaptation technique that uses style perturbation from a single image to enhance object detection in different target domains.

Findings

OSSA achieves state-of-the-art results among one-shot domain adaptation methods.

In some cases, OSSA surpasses methods using thousands of unlabeled images.

Effective in weather, sim2real, and thermal adaptation scenarios.

Abstract

Despite their success in various vision tasks, deep neural network architectures often underperform in out-of-distribution scenarios due to the difference between training and target domain style. To address this limitation, we introduce One-Shot Style Adaptation (OSSA), a novel unsupervised domain adaptation method for object detection that utilizes a single, unlabeled target image to approximate the target domain style. Specifically, OSSA generates diverse target styles by perturbing the style statistics derived from a single target image and then applies these styles to a labeled source dataset at the feature level using Adaptive Instance Normalization (AdaIN). Extensive experiments show that OSSA establishes a new state-of-the-art among one-shot domain adaptation methods by a significant margin, and in some cases, even outperforms strong baselines that use thousands of unlabeled target images. By applying OSSA in various scenarios, including weather, simulated-to-real (sim2real), and visual-to-thermal adaptations, our study explores the overarching significance of the style gap in these contexts. OSSA's simplicity and efficiency allow easy integration into existing frameworks, providing a potentially viable solution for practical applications with limited data availability. Code is available at https://github.com/RobinGerster7/OSSA