Research map / 3D vision / Human pose estimation
Populating 3D Scenes by Learning Human-Scene Interaction
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 |
|---|---|---|---|
| Add scene semantic features to contact features (ablation) | POSA scene population · PROX test set vs. PROX GT | Generation preference (%): 46.9% -> 49.1% | Helped |
| Add scene semantic features to contact features (ablation) | POSA scene population · Human-scene generation vs. PLACE | Preference for POSA variant (%): 60.7% -> 61.0% | Helped |
| Use clothed meshes with semantic features instead of contact-only features (ablation) | POSA-clothing · PROX test set vs. PROX GT | Generation preference (%): 55.0% -> 60.6% | Helped |
| Replace hand-selected PROX contact points with learned POSA interaction features | Monocular pose estimation · PROX | PJE (mm): 167.08 -> 154.33 | Helped |
| Replace hand-selected PROX contact points with learned POSA interaction features | Monocular pose estimation · PROX | V2V (mm): 166.51 -> 154.84 | Helped |
| Replace hand-selected PROX contact points with learned POSA interaction features | Monocular pose estimation · PROX | p.PJE (mm): 71.97 -> 73.17 | Hurt |
| Replace hand-selected PROX contact points with learned POSA interaction features | Monocular pose estimation · PROX | p.V2V (mm): 61.14 -> 63.23 | Hurt |
| Add scene semantics to contact-only features (ablation) | POSA scene population · PROX test scenes | Non-collision score: 0.97 -> 0.97 | No clear effect |
| Add scene semantics to contact-only features (ablation) | POSA scene population · PROX test scenes | Contact score: 1.0 -> 0.99 | Hurt |
| Add scene semantics to contact-only features (ablation) | POSA scene population · PROX test scenes | Entropy: 2.94 -> 2.92 | Hurt |
| Add scene semantics to contact-only features (ablation) | POSA scene population · PROX test scenes | Cluster size: 2.28 -> 2.27 | Hurt |
Closest papers by meaning
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.90)
- Reconstructing Close Human Interaction with Appearance and Proxemics Reasoning (CVPR 2025 · similarity 0.88)
- PICO: Reconstructing 3D People In Contact with Objects (CVPR 2025 · similarity 0.88)
- GenZI: Zero-Shot 3D Human-Scene Interaction Generation (CVPR 2024 · similarity 0.88)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (CVPR 2025 · similarity 0.88)
- InteractVLM: 3D Interaction Reasoning from 2D Foundational Models (CVPR 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.