FedStyle: Style-Based Federated Learning Crowdsourcing Framework for Art Commissions

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ran, Changjuan, Guo, Yeting, Liu, Fang, Cui, Shenglan, Ye, Yunfan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913329090396160
author Ran, Changjuan
Guo, Yeting
Liu, Fang
Cui, Shenglan
Ye, Yunfan
author_facet Ran, Changjuan
Guo, Yeting
Liu, Fang
Cui, Shenglan
Ye, Yunfan
contents The unique artistic style is crucial to artists' occupational competitiveness, yet prevailing Art Commission Platforms rarely support style-based retrieval. Meanwhile, the fast-growing generative AI techniques aggravate artists' concerns about releasing personal artworks to public platforms. To achieve artistic style-based retrieval without exposing personal artworks, we propose FedStyle, a style-based federated learning crowdsourcing framework. It allows artists to train local style models and share model parameters rather than artworks for collaboration. However, most artists possess a unique artistic style, resulting in severe model drift among them. FedStyle addresses such extreme data heterogeneity by having artists learn their abstract style representations and align with the server, rather than merely aggregating model parameters lacking semantics. Besides, we introduce contrastive learning to meticulously construct the style representation space, pulling artworks with similar styles closer and keeping different ones apart in the embedding space. Extensive experiments on the proposed datasets demonstrate the superiority of FedStyle.
format Preprint
id arxiv_https___arxiv_org_abs_2404_16336
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FedStyle: Style-Based Federated Learning Crowdsourcing Framework for Art Commissions
Ran, Changjuan
Guo, Yeting
Liu, Fang
Cui, Shenglan
Ye, Yunfan
Machine Learning
Computer Vision and Pattern Recognition
The unique artistic style is crucial to artists' occupational competitiveness, yet prevailing Art Commission Platforms rarely support style-based retrieval. Meanwhile, the fast-growing generative AI techniques aggravate artists' concerns about releasing personal artworks to public platforms. To achieve artistic style-based retrieval without exposing personal artworks, we propose FedStyle, a style-based federated learning crowdsourcing framework. It allows artists to train local style models and share model parameters rather than artworks for collaboration. However, most artists possess a unique artistic style, resulting in severe model drift among them. FedStyle addresses such extreme data heterogeneity by having artists learn their abstract style representations and align with the server, rather than merely aggregating model parameters lacking semantics. Besides, we introduce contrastive learning to meticulously construct the style representation space, pulling artworks with similar styles closer and keeping different ones apart in the embedding space. Extensive experiments on the proposed datasets demonstrate the superiority of FedStyle.
title FedStyle: Style-Based Federated Learning Crowdsourcing Framework for Art Commissions
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.16336