Skip to content
Free US shipping over $75 · 30-day guarantee
AI Generation

Style Transfer

Style transfer is a technique that applies the visual style of one image (like a painting) to the content of another (like a photograph).

Style transfer originated in 2015 as a neural-network technique for decomposing images into "content" and "style" and then recombining them: take the content of a photograph and render it in the style of a Van Gogh painting.

Modern diffusion-based workflows accomplish similar results through image-to-image generation with style-specific prompts, often combined with reference-image conditioning. The results tend to be more coherent than classical style transfer, with fewer of the telltale "neural painting" artifacts.

Style transfer is ethically sensitive: applying a specific living artist's style commercially is generally considered bad practice, while applying historical or broadly-defined styles (impressionism, Art Nouveau, Bauhaus) is well-accepted.

Related terms

Ready to make something?

Describe what you want, or pick a style. We’ll preview it in your room before you order.

Start creating