Research map / 3D vision / 3D generation
Generating Human Interaction Motions in Scenes with Text Control
Hongwei Yi, Justus Thies, Michael J. Black, Xue Bin Peng, Davis Rempe
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 |
|---|---|---|---|
| pre-train scene-agnostic branch then fine-tune scene-aware branch instead of 1-stage training (ablation) | TeSMo navigation · Loco-3D-FRONT | Root position error (m): 0.197 -> 0.169 | Improved |
| pre-train then fine-tune instead of 1-stage training (ablation) | TeSMo navigation · Loco-3D-FRONT | R-precision (top-3): 0.152 -> 0.376 | Improved |
| use two-branch architecture instead of single-branch TRACE-style floor map input (ablation) | TeSMo navigation · Loco-3D-FRONT | Collision ratio: 0.055 -> 0.031 | Improved |
| two-stage pelvis trajectory then in-painting instead of one-stage full-body generation (ablation) | TeSMo navigation · Loco-3D-FRONT | Root position error (m): 0.844 -> 0.169 | Improved |
| use PriorMDM in-painting instead of OmniControl in-painting (ablation) | TeSMo navigation · Loco-3D-FRONT | Root position error (m): 0.459 -> 0.169 | Improved |
| use OmniControl in-painting instead of PriorMDM (ablation) | TeSMo navigation · Loco-3D-FRONT | FID (full-body): 20.465 -> 17.927 | Improved |
| add goal-reaching test-time guidance (ablation) | TeSMo navigation · Loco-3D-FRONT | Goal position error: 0.1568 -> 0.118 | Improved |
| add collision test-time guidance (ablation) | TeSMo navigation · Loco-3D-FRONT | Collision ratio: 0.0294 -> 0.0013 | Improved |
| add goal-reaching guidance without collision guidance (ablation) | TeSMo navigation · Loco-3D-FRONT | Collision ratio: 0.0294 -> 0.0342 | Hurt |
| add collision guidance to interaction model (ablation) | TeSMo interaction · SAMP sitting | Penetration ratio: 0.0611 -> 0.0414 | Improved |
| add goal-reaching guidance to interaction model (ablation) | TeSMo interaction · SAMP sitting | Goal position error: 0.1445 -> 0.1453 | No clear effect |
| use TeSMo diffusion model instead of RL-based DIMOS | TeSMo interaction · SAMP sitting | Goal position error: 0.2020 -> 0.1445 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TRACE — TRACE: 5D Temporal Regression of Avatars With Dynamic Cameras in 3D Environments (CVPR 2023) · 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
- Generating Human Motion in 3D Scenes from Text Descriptions (CVPR 2024 · similarity 0.90)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.89)
- Auto-Regressive Diffusion for Generating 3D Human-Object Interactions (AAAI 2025 · similarity 0.88)
- HOIAnimator: Generating Text-prompt Human-object Animations using Novel Perceptive Diffusion Models (CVPR 2024 · similarity 0.88)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.88)
- HSI-GPT: A General-Purpose Large Scene-Motion-Language Model for Human Scene Interaction (CVPR 2025 · similarity 0.88)
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.
