Research map / 3D vision / 3D generation
PAT3D: Physics-Augmented Text-to-3D Scene Generation
Guying Lin, Kemeng Huang, Michael Liu, Ruihan Gao, Hanke Chen, Lyuhao Chen, Beijia Lu, Taku Komura, Yuan Liu, Jun-Yan Zhu, Minchen Li
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add scene-tree layout initialization and differentiable-simulation optimization to the raw depth layout (ablation) | PAT3D · PAT3D test prompts (18) | Physical plausibility score: 65.5 -> 88.5 | Improved |
| add simulation-in-the-loop optimization after scene initialization (ablation) | PAT3D · PAT3D test prompts (18) | Simulated displacement: 2.91 -> 0 | Improved |
| add simulation-in-the-loop optimization after scene initialization (ablation) | PAT3D · PAT3D test prompts (18) | Physical plausibility score: 34.2 -> 88.5 | Improved |
| add scene-tree layout initialization to raw layout (ablation) | PAT3D · PAT3D test prompts (18) | Penetration ratio: 14.11 -> 0 | Improved |
| add scene-tree layout initialization to raw layout (without optimization) (ablation) | PAT3D · PAT3D test prompts (18) | Physical plausibility score: 65.5 -> 34.2 | Hurt |
| add layout initialization and optimization on top of raw layout (ablation) | PAT3D · PAT3D test prompts (18) | CLIP Score: 29.88 -> 31.79 | Improved |
| full PAT3D vs Scene Init. only (ablation) | PAT3D · PAT3D test prompts (18) | VQA Score: 0.70 -> 0.68 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MIDI — MIDI: Multi-Instance Diffusion for Single Image to 3D Scene Generation (CVPR 2025) · better on 5 of 5 shared results
- GraphDreamer — GraphDreamer: Compositional 3D Scene Synthesis from Scene Graphs (CVPR 2024) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PhysGen3D: Crafting a Miniature Interactive World from a Single Image (CVPR 2025 · similarity 0.88)
- 3D-SceneDreamer: Text-Driven 3D-Consistent Scene Generation (CVPR 2024 · similarity 0.87)
- GALA3D: Towards Text-to-3D Complex Scene Generation via Layout-guided Generative Gaussian Splatting (ICML 2024 · similarity 0.87)
- PhysX-3D: Physical-Grounded 3D Asset Generation (NeurIPS 2025 · similarity 0.87)
- Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication (NeurIPS 2024 · similarity 0.87)
- Simulation-Ready Compositional 3D Scene Reconstruction from a Single Image (NeurIPS 2026 · similarity 0.87)
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.
