English

Generative Phomosaic with Structure-Aligned and Personalized Diffusion

Computer Vision and Pattern Recognition 2026-04-09 v1 Artificial Intelligence

Abstract

We present the first generative approach to photomosaic creation. Traditional photomosaic methods rely on a large number of tile images and color-based matching, which limits both diversity and structural consistency. Our generative photomosaic framework synthesizes tile images using diffusion-based generation conditioned on reference images. A low-frequency conditioned diffusion mechanism aligns global structure while preserving prompt-driven details. This generative formulation enables photomosaic composition that is both semantically expressive and structurally coherent, effectively overcoming the fundamental limitations of matching-based approaches. By leveraging few-shot personalized diffusion, our model is able to produce user-specific or stylistically consistent tiles without requiring an extensive collection of images.

Keywords

Cite

@article{arxiv.2604.06989,
  title  = {Generative Phomosaic with Structure-Aligned and Personalized Diffusion},
  author = {Jaeyoung Chung and Hyunjin Son and Kyoung Mu Lee},
  journal= {arXiv preprint arXiv:2604.06989},
  year   = {2026}
}

Comments

Project page: https://robot0321.github.io/GenerativePhotomosaic/index.html

R2 v1 2026-07-01T11:59:08.839Z