Research map / 3D vision / 3D generation
RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces
Sebastien Ehrhardt, Oliver Groth, Aron Monszpart, Martin Engelcke, Ingmar Posner, Niloy Mitra, Andrea Vedaldi
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 spatial interaction module and position regression loss (vs BlockGAN2D) (ablation) | RELATE · Balls-in-bowl | disentanglement / FID: 98.0 -> 26.0 disentanglement distance | Improved |
| add residual connection in correction network (ablation) | RELATE · Balls-in-bowl | FID / disentanglement | Improved |
| add position regression loss (ablation) | RELATE · Balls-in-bowl | FID / disentanglement | Improved |
| model object correlations with interaction module (vs BlockGAN2D) (ablation) | RELATE · CLEVR | FID: 78.1 -> 62.9 | Improved |
| model object correlations with interaction module (vs BlockGAN2D) (ablation) | RELATE · ShapeStacks | disentanglement distance: 272.0 -> 17.0 | Improved |
| add scale prediction for each object | RELATE · CLEVR | FID: 62.9 -> 44.9 | Improved |
| add scale prediction for each object | RELATE · Cars | FID: 42.0 -> 46.8 | Hurt |
| add scale prediction for each object | RELATE · Cars video | FVD: 2253 -> 1855 | Improved |
| add scale prediction for each object | RELATE · CLEVR-5vbg | FID: 36.4 -> 35.9 | No clear effect |
Closest papers by meaning
- Exploiting Relationship for Complex-scene Image Generation (AAAI 2021 · similarity 0.90)
- Compositional Scene Understanding through Inverse Generative Modeling (ICML 2025 · similarity 0.88)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · similarity 0.87)
- DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videos (NeurIPS 2024 · similarity 0.87)
- From Programs to Poses: Factored Real-World Scene Generation via Learned Program Libraries (NeurIPS 2025 · similarity 0.87)
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · 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.
