Discovering and Personalizing Artistic Styles with Generative Models

Text-to-image models have gained widespread popularity, transforming digital art creation by allowing users to generate highly detailed and imaginative visual content from natural language prompts. These models are now widely adopted across various domains, particularly in the arts, where they enable a broad range of creative expression and make artistic creation more accessible to a wider audience. In this work, we focus on discovering and personalizing emergent artistic styles. By clustering millions of user-generated images from Artbreeder—a platform with over 13 million users—we uncover a rich landscape of previously undocumented unique styles, transcending conventional categories like 'cyberpunk' or 'Picasso,' that reflect the collective creative exploration of users worldwide. Building on these discoveries, we assess personalization methods to align generated content with individual aesthetic preferences and introduce a style recommendation system based on historical behavior. To support this effort, we curate and release STYLEBREEDER, a large-scale dataset comprising 6.8 million images and 1.8 million text prompts from 95,000 users, along with clustering annotations and stylistic embeddings. We also introduce the Style Atlas platform, providing public access to 100 curated style LoRA models for user experimentation. Our work demonstrates that AI is not only a tool for generating art, but also a means of discovering and fostering emerging forms of creativity. By openly sharing our data, models, and tools, we aim to support a more diverse, inclusive, and collaborative digital art ecosystem. All resources are available at https://stylebreeder.github.io under a Public Domain (CC0) license.

Name thesis:degree_name

Master of Science

Level thesis:degree_level

masters

Discipline thesis:degree_discipline

Computer Science & Applications

Department dc:contributor.department

Computer Science and#38; Applications

Grantor dc:publisher

Virginia Tech

Year dc:date.issued

2025

Author dc:creator

• Zheng, Matthew

Chair dc:contributor.committeechair

• Yanardag Delul, Pinar

Committee members dc:contributor.committeemember

• Eldardiry, Hoda Mohamed
• Thomas, Christopher Lee

• Computer Vision
• Generative Models
• Text-to-Image Generation
• Artificial Intelligence

Statement dc:rights

• In Copyright

Licence dc:rights.uri

• http://rightsstatements.org/vocab/InC/1.0/

Language dc:language.iso

en

Dc Identifier Other

vt_gsexam:44223

OAI identifier oai:identifier

oai:vtechworks.lib.vt.edu:10919/135576