Generative Data Augmentation for Few-Shot Domain Adaptation

Carlos López Fortín, Ikuko Nishikawa

2024

Abstract

Domain adaptation in computer vision focuses on addressing the domain gap between source and target distributions, generally via adversarial methods or feature distribution alignment. However, most of them suppose the availability of sufficient target data to properly teach the model domain-invariant representations. Few-shot scenarios where target data is scarce pose a significant challenge for their implementation in real-world scenarios. Leveraging fine-tuned diffusion models for synthetic data augmentation, we present Generative Data Augmentation for Few-shot Domain Adaptation, a model-agnostic approach to address the Few-shot problem in domain adaptation for multi-class classification. Experimental results show that using augmented data from fine-tuned diffusion models with open-source data sets can improve average accuracy by up to 3%, as well as increase per-class accuracy between 3% to 30%, for state-of-the-art domain adaptation methods with respect to their non-augmented counterparts, without requiring any major modifications to their architecture. This provides an easy-to-implement solution for the adoption of domain adaptation methods in practical scenarios.

Paper Citation

in Harvard Style

López Fortín C. and Nishikawa I. (2024). Generative Data Augmentation for Few-Shot Domain Adaptation. In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-684-2, SciTePress, pages 256-265. DOI: 10.5220/0012338000003654

in Bibtex Style

@conference{icpram24,
author={Carlos López Fortín and Ikuko Nishikawa},
title={Generative Data Augmentation for Few-Shot Domain Adaptation},
booktitle={Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2024},
pages={256-265},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012338000003654},
isbn={978-989-758-684-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Generative Data Augmentation for Few-Shot Domain Adaptation
SN - 978-989-758-684-2
AU - López Fortín C.
AU - Nishikawa I.
PY - 2024
SP - 256
EP - 265
DO - 10.5220/0012338000003654
PB - SciTePress