Generative Phomosaic with Structure-Aligned and Personalized Diffusion
This addresses the need for more diverse and personalized photomosaics in creative applications, representing a novel paradigm shift rather than an incremental improvement.
The paper tackles the problem of photomosaic creation by introducing a generative approach using diffusion models, which overcomes limitations of traditional matching-based methods by enabling semantically expressive and structurally coherent compositions with few-shot personalization.
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.