Research map / 3D vision / 3D generation
Towards Diverse and Natural Scene-Aware 3D Human Motion Synthesis
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the full hierarchical framework | Scene-aware motion synthesis · PROX | FD: 126.46 -> 112.74 | Helped |
| Use the full hierarchical framework | Scene-aware motion synthesis · PROX | APD: 2.46 -> 2.77 | Helped |
| Add the Path Refiner (ablation) | Motion completion network · PROX | FD: 126.46 -> 112.74 | Helped |
| Add the Path Refiner (ablation) | Motion completion network · Matterport3D | APD: 2.34 -> 2.57 | Helped |
| Add pose-related sampling to disperse anchors sharing an action (ablation) | Anchor placement · PROX | Anchor entropy: 2.62 -> 2.74 | Helped |
| Add the Place Refiner (ablation) | Anchor placement with pose-related sampling · PROX | Anchor entropy: 2.74 -> 2.77 | Helped |
| Use a learned Neural Mapper instead of standard A* | Path planning · PROX planning | 1/2 path-distance STD: 0 -> 0.508 | Helped |
| Use a learned Neural Mapper instead of random noise (ablation) | Path planning · PROX planning | User Study: 2.52(0.53) -> 4.27(0.52) | Helped |
| Add random noise to A* path costs (ablation) | Path planning · PROX planning | User Study: 4.31(0.48) -> 2.52(0.53) | Hurt |
| Add the optimization post-process (ablation) | Synthesized motions · PROX | Non-Collision: 95.93 -> 99.61 | Helped |
| Include both translation and orientation from the planned path in positional encoding (ablation) | Path Refiner · PROX | Naturalness | Helped |
Closest papers by meaning
- Harmonizing Stochasticity and Determinism: Scene-responsive Diverse Human Motion Prediction (NeurIPS 2024 · similarity 0.89)
- Deterministic-to-Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis (CVPR 2025 · similarity 0.87)
- Motion Diversification Networks (CVPR 2024 · similarity 0.87)
- ONE-SHOT: Compositional Human-Environment Video Synthesis via Spatial-Decoupled Motion Injection and Hybrid Context Integration (NeurIPS 2026 · similarity 0.87)
- Sitcom-Crafter: A Plot-Driven Human Motion Generation System in 3D Scenes (ICLR 2025 · similarity 0.86)
- Move as You Say Interact as You Can: Language-guided Human Motion Generation with Scene Affordance (CVPR 2024 · similarity 0.86)
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.