A trainable adapter steers a frozen Stable Diffusion model with semantic 3D bounding boxes and a camera pose, producing images that respect 3D layout, viewpoint, and per-object captions.
Scaling recti- fied flow transformers for high-resolution image synthesis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LACONIC: A 3D Layout Adapter for Controllable Image Creation
A trainable adapter steers a frozen Stable Diffusion model with semantic 3D bounding boxes and a camera pose, producing images that respect 3D layout, viewpoint, and per-object captions.