Research map / 3D vision / 3D generation
Grounded Latents for Entity-Centric 4D Scene Generation
Jinhyung Park, Navyata Sanghvi, Erica Weng, Shawn Hunt, Shinya Tanaka, Hironobu Fujiyoshi, Kris Kitani
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 |
|---|---|---|---|
| increase number of latents from 256 to 768 (ablation) | LatentWorld · CarlaSC | geometry MMD: 13.32 -> 6.44 | Improved |
| increase number of latents from 768 to 1024 (ablation) | LatentWorld · CarlaSC | semantics MMD: 11.30 -> 12.36 | Hurt |
| increase number of latents from 768 to 1024 (ablation) | LatentWorld · CarlaSC | reconstruction mIoU: 93.63 -> 94.71 | Improved |
| increase number of latents from 768 to 1024 (ablation) | LatentWorld · CarlaSC | geometry+semantics MMD: 6.69 -> 7.30 | Hurt |
| add outpainting mean-shift guidance (lambda 1.0) (ablation) | LatentWorld outpainting · Waymo | geometry+semantics MMD: 2.31 -> 1.04 | Improved |
| increase outpainting push weight from 1.0 to 1.5 (ablation) | LatentWorld outpainting · Waymo | geometry+semantics MMD: 1.04 -> 1.09 | Hurt |
| replace dense voxel/grid generation with grounded 3D latents | DynamicCity · Waymo 4D | geometry+semantics MMD: 2.61 -> 0.96 | Improved |
| replace grid-based generators with grounded latents on CarlaSC | SemCity · CarlaSC | geometry MMD: 10.47 -> 6.44 | Improved |
| grounded latents vs grid baselines on semantics-only metric | PDD/SemCity/DynamicCity · CarlaSC | semantics MMD: 12.74 -> 11.30 | 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.
- DynamicCity — DynamicCity: Large-Scale 4D Occupancy Generation from Dynamic Scenes (ICLR 2025) · better on 6 of 6 shared results
- SemCity — SemCity: Semantic Scene Generation with Triplane Diffusion (CVPR 2024) · better on 4 of 4 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
- 4DWeaver: Bridging Reconstruction and Generation Via Compact Autoregressive Priors (TPAMI 2026 · similarity 0.90)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · similarity 0.89)
- LT3SD: Latent Trees for 3D Scene Diffusion (CVPR 2025 · similarity 0.88)
- Controlling Space and Time with Diffusion Models (ICLR 2025 · similarity 0.87)
- 4Real: Towards Photorealistic 4D Scene Generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.87)
- A Unified Approach for Text- and Image-guided 4D Scene Generation (CVPR 2024 · 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.
