Research map / 3D vision / 3D generation
SceneTransporter: Optimal Transport-Guided Compositional Latent Diffusion for Single-Image Structured 3D Scene Generation
Ling Wang, Hao-Xiang Guo, Xinzhou Wang, Fuchun Sun, Kai Sun, Pengkun Liu, Hang Xiao, Zhong Wang, Guangyuan Fu, Eric Li, Yang Liu, Yikai Wang
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 OT plan-gated cross-attention (ablation) | PartPacker · open-world scenes | IoU mean: 0.2142 -> 0.0926 | Improved |
| add OT plan-gated cross-attention (ablation) | PartPacker · open-world scenes | Uni3D: 0.2887 -> 0.3021 | Improved |
| add edge-regularized assignment cost (ablation) | SceneTransporter · open-world scenes | IoU max: 0.0241 -> 0.0101 | Improved |
| add edge-regularized assignment cost (ablation) | SceneTransporter · open-world scenes | IoU mean: 0.1136 -> 0.0926 | Improved |
| add edge-regularized assignment cost (ablation) | SceneTransporter · open-world scenes | inference time (s): 54.61 -> 54.99 | No clear effect |
| OT gating adds compute to PartPacker | SceneTransporter · open-world scenes | inference time (s): 47.41 -> 54.99 | Hurt |
| raise entropic regularization from 0.10 to 0.12 (ablation) | SceneTransporter · open-world scenes | IoU max: 0.0101 -> 0.1239 | Hurt |
| lower edge sensitivity gamma_edge to 6.0 (ablation) | SceneTransporter · open-world scenes | ULIP: 0.1466 -> 0.1488; IoU max 0.0101 -> 0.0243 | Mixed |
| raise edge sensitivity gamma_edge to 12.0 (ablation) | SceneTransporter · open-world scenes | IoU max: 0.0101 -> 0.1149 | Hurt |
| change edge smoothing lambda_edge from 0.8 to 1.0 (ablation) | SceneTransporter · open-world scenes | IoU max: 0.0101 -> 0.1482 | Hurt |
| apply OT gating in 1/3 instead of 1/2 of DiT blocks (ablation) | SceneTransporter · open-world scenes | IoU max: 0.0101 -> 0.0992 | Hurt |
| apply OT gating in all DiT blocks instead of half (ablation) | SceneTransporter · open-world scenes | ULIP-2: 0.3220 -> 0.3262; time 54.99 -> 65.24 | 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 6 of 6 shared results
- PartCrafter — PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025) · better on 4 of 6 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
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · similarity 0.91)
- LT3SD: Latent Trees for 3D Scene Diffusion (CVPR 2025 · similarity 0.90)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · similarity 0.89)
- DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024 · similarity 0.89)
- Move Anything with Layered Scene Diffusion (CVPR 2024 · similarity 0.89)
- Coherent 3D Scene Diffusion From a Single RGB Image (NeurIPS 2024 · similarity 0.89)
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.
