Compositional Visual Generation with Composable Diffusion Models (original) (raw)

Compositional Generation using Stable Diffusion. Our proposed Conjunction (AND) and Negation (NOT) can be applied to conditional diffusion models for compositional generation. Our operators are added into Stable Diffusion WebUI! Corresponding pages are as follows: Conjunction (AND) andNegation (NOT).

"mystical trees" AND "A magical pond" AND "Dark"

"mystical trees" AND "A magical pond" AND NOT "Dark"

"A stone castle surrounded by lakes and trees, fantasy, wallpaper, concept art, extremely detailed" AND "Black and white"

"A stone castle surrounded by lakes and trees, fantasy, wallpaper, concept art AND extremely detailed" AND NOT "Black and white"