AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

LayoutGPT: Compositional Visual Planning and Generation with Large Language Models

Weixi Feng, Wanrong Zhu, Tsu-Jui Fu, Varun Jampani, Arjun Reddy Akula, Xuehai He, S Basu, Xin Eric Wang, William Yang Wang

NeurIPS 2023 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 62 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Represent layouts as CSS-style structured programs rather than comma-separated values (ablation)LayoutGPT · NSR-1K spatial reasoning promptslayout accuracy: 55.12 -> 80.82Improved
Add task instructions (ablation)LayoutGPT · NSR-1K spatial reasoning promptslayout accuracy: 55.12 -> 78.23Improved
Add normalization when used alone (ablation)LayoutGPT · NSR-1K spatial reasoning promptslayout accuracy: 55.12 -> 44.10Hurt
Add normalization to task instructions and CSS formatting (ablation)LayoutGPT · NSR-1K spatial reasoning promptslayout accuracy: 81.84 -> 82.54Improved
Add normalization to the 3D layout prompt (ablation)LayoutGPT · 3D-FRONT bedroom testout-of-bounds rate: 55.32 -> 46.57Improved
Add CSS structure to the 3D layout prompt (ablation)LayoutGPT · 3D-FRONT bedroom testout-of-bounds rate: 55.32 -> 51.77Improved
Use retrieved exemplars similar to the test room's floor plan instead of random bedroom exemplars (ablation)LayoutGPT · 3D-FRONT bedroom testout-of-bounds rate: 85.58 -> 43.26Improved
Generate layouts with LayoutGPT instead of LayoutTransformer2D layout planning · NSR-1K numerical reasoninglayout accuracy: 22.26 -> 86.33Improved
Use a layout-planning stage with LayoutGPT rather than direct Stable Diffusion text-to-image generationNumerical image generation · NSR-1K numerical reasoningGLIP image accuracy: 42.44 -> 55.64Improved
Use LayoutGPT layouts rather than supervised ATISS layouts3D indoor scene synthesis · 3D-FRONT bedroom testKL divergence: 0.0113 -> 0.0995Hurt
Use LayoutGPT instead of ATISS3D indoor scene synthesis · 3D-FRONT living room testFID: 85.40 -> 76.34Improved
Use LayoutGPT + ReCo rather than Attend-and-ExciteColor attribute binding · HRS-Bench color prompts, overallattribute binding accuracy: 22.96 -> 36.96Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy