Single-Shot Domain Adaptation via Target-Aware Generative Augmentation

TL;DR

This paper introduces SiSTA, a novel single-shot domain adaptation method that fine-tunes generative models with minimal target data to improve model generalization across large distribution shifts.

Contribution

The paper proposes SiSTA, a new approach that fine-tunes generative models with a single target sample and employs novel sampling for synthetic data, enhancing adaptation performance.

Findings

SiSTA improves adaptation accuracy by up to 20% over baselines.

It performs comparably to models trained on larger target datasets.

Effective in face attribute detection under significant domain shifts.

Abstract

The problem of adapting models from a source domain using data from any target domain of interest has gained prominence, thanks to the brittle generalization in deep neural networks. While several test-time adaptation techniques have emerged, they typically rely on synthetic data augmentations in cases of limited target data availability. In this paper, we consider the challenging setting of single-shot adaptation and explore the design of augmentation strategies. We argue that augmentations utilized by existing methods are insufficient to handle large distribution shifts, and hence propose a new approach SiSTA (Single-Shot Target Augmentations), which first fine-tunes a generative model from the source domain using a single-shot target, and then employs novel sampling strategies for curating synthetic target data. Using experiments with a state-of-the-art domain adaptation method, we find that SiSTA produces improvements as high as 20\% over existing baselines under challenging shifts in face attribute detection, and that it performs competitively to oracle models obtained by training on a larger target dataset.