Research map / 3D vision / 3D generation
IntentMotion: Learning Intent-Aware Human Motion from Language in 3D Scenes
Wenfeng Song, Shi Zheng, Xinyu Zhang, Xingliang Jin, Aimin Hao, Fei Hou, Xia Hou, Shuai Li
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 contact conditioning (IGCF) (ablation) | IntentMotion · HUMANISE | FID: 0.233 -> 0.163 | Improved |
| add hierarchical attention instead of MLP for contact fusion (ablation) | IGCF · HUMANISE | FID: 0.203 -> 0.163 | Improved |
| increase alignment depth from 4 to 6 self-attention layers (ablation) | IGCF · HUMANISE | FID: 0.163 -> 0.216 | Hurt |
| reduce alignment depth from 4 to 2 (ablation) | IGCF · HUMANISE | FID: 0.163 -> 0.156 | No clear effect |
| add coarse and fine guidance to Adapt-Dec (ablation) | IADM · HUMANISE | FID: 0.200 -> 0.163 | Improved |
| use Adapt-Dec instead of U-Net denoiser (ablation) | IADM · HUMANISE | FID: 0.994 -> 0.163 | Improved |
| use Adapt-Dec instead of DiT denoiser (ablation) | IADM · HUMANISE | FID: 0.247 -> 0.163 | Improved |
| add IGCF module to Adapt-Dec baseline (ablation) | Cen et al.* · LINGO | penetration rate: 0.261 -> 0.119 | Improved |
| add target object grounding | IntentMotion · HUMANISE | contact: 95.63 -> 95.30 contact; FID 0.255 -> 0.163 | Mixed |
Closest papers by meaning
- Move as You Say Interact as You Can: Language-guided Human Motion Generation with Scene Affordance (CVPR 2024 · similarity 0.88)
- CG-HOI: Contact-Guided 3D Human-Object Interaction Generation (CVPR 2024 · similarity 0.88)
- HSI-GPT: A General-Purpose Large Scene-Motion-Language Model for Human Scene Interaction (CVPR 2025 · similarity 0.88)
- The Language of Motion: Unifying Verbal and Non-verbal Language of 3D Human Motion (CVPR 2025 · similarity 0.87)
- Generating Human Motion in 3D Scenes from Text Descriptions (CVPR 2024 · similarity 0.87)
- HOIAnimator: Generating Text-prompt Human-object Animations using Novel Perceptive Diffusion Models (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.
