Salvato in:
Dettagli Bibliografici
Autori principali: Loures, Clarissa, Hosken, Caio, Oliveira, Luan, Zuin, Gianlucca, Veloso, Adriano
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2603.04343
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910041131450368
author Loures, Clarissa
Hosken, Caio
Oliveira, Luan
Zuin, Gianlucca
Veloso, Adriano
author_facet Loures, Clarissa
Hosken, Caio
Oliveira, Luan
Zuin, Gianlucca
Veloso, Adriano
contents Attributing authorship to paintings is a historically complex task, and one of its main challenges is the limited availability of real artworks for training computational models. This study investigates whether synthetic images, generated through DreamBooth fine-tuning of Stable Diffusion, can improve the performance of classification models in this context. We propose a hybrid approach that combines real and synthetic data to enhance model accuracy and generalization across similar artistic styles. Experimental results show that adding synthetic images leads to higher ROC-AUC and accuracy compared to using only real paintings. By integrating generative and discriminative methods, this work contributes to the development of computer vision techniques for artwork authentication in data-scarce scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2603_04343
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Enhancing Authorship Attribution with Synthetic Paintings
Loures, Clarissa
Hosken, Caio
Oliveira, Luan
Zuin, Gianlucca
Veloso, Adriano
Computer Vision and Pattern Recognition
Machine Learning
Attributing authorship to paintings is a historically complex task, and one of its main challenges is the limited availability of real artworks for training computational models. This study investigates whether synthetic images, generated through DreamBooth fine-tuning of Stable Diffusion, can improve the performance of classification models in this context. We propose a hybrid approach that combines real and synthetic data to enhance model accuracy and generalization across similar artistic styles. Experimental results show that adding synthetic images leads to higher ROC-AUC and accuracy compared to using only real paintings. By integrating generative and discriminative methods, this work contributes to the development of computer vision techniques for artwork authentication in data-scarce scenarios.
title Enhancing Authorship Attribution with Synthetic Paintings
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2603.04343