Research map / Video / Video generation
PINO: Person-Interaction Noise Optimization for Long-Duration and Customizable Motion Generation of Arbitrary-Sized Groups
Sakuya Ota, Qing Yu, Kent Fujiwara, Satoshi Ikehata, Ikuro Sato
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 PINO overlap-penalty noise optimization | InterGen · Two-person interaction (InterHuman, 300 samples) | Overlap: 0.119 -> 0.000 | Improved |
| add PINO noise optimization | InterGen · Two-person interaction (InterHuman, 300 samples) | Penetration volume (cm^3): 3112.72 -> 275.65 | Improved |
| add PINO noise optimization | InterGen · Two-person interaction (InterHuman, 300 samples) | FID: 13.278 -> 13.163 | Improved |
| add PINO to sequential multi-person generation | InterGen · Five-person generation, pair (1,5) | Overlap: 0.977 -> 0.069 | Improved |
| add PINO to sequential multi-person generation | InterGen · Five-person generation, pair (1,5) | Foot skate: 0.124 -> 0.142 | Hurt |
| add root-position penalty (ablation) | InterGen · Penalty ablation (two agents, 24 sequences per configuration) | Position error rate: 1.0 -> 0.0 | Improved |
| add root-position penalty (ablation) | InterGen · Penalty ablation (two agents, 24 sequences per configuration) | Region violation rate: 0.500 -> 0.917 | Hurt |
| add overlap penalty (ablation) | InterGen + root penalty · Penalty ablation (two agents, 24 sequences per configuration) | Overlap rate: 0.333 -> 0.0 | Improved |
| add movement region penalty (ablation) | InterGen + root + overlap · Penalty ablation (two agents, 24 sequences per configuration) | Region violation rate: 0.958 -> 0.043 | Improved |
| add orientation penalty (ablation) | InterGen + root + overlap + region · Penalty ablation (two agents, 24 sequences per configuration) | Orientation error rate: 1.0 -> 0.208 | Improved |
| add PINO to motion extension | InterGen (inpainting) · Motion extension (two-person, non-semantic) | Overlap: 0.672 -> 0.031 | Improved |
| add PINO to motion extension | InterGen · Motion extension (two-person, cropped to GT length) | FID: 20.390 -> 18.930 | Improved |
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.
- InterGen — InterGen: Diffusion-Based Multi-human Motion Generation Under Complex Interactions (IJCV 2024) · better on 11 of 13 shared results
- FreeMotion — FreeMotion: A Unified Framework for Number-free Text-to-Motion Synthesis (ECCV 2024) · better on 2 of 2 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
- InterControl: Zero-shot Human Interaction Generation by Controlling Every Joint (NeurIPS 2024 · similarity 0.88)
- Ready-to-React: Online Reaction Policy for Two-Character Interaction Generation (ICLR 2025 · similarity 0.86)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.86)
- CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects (NeurIPS 2025 · similarity 0.85)
- Learning to Generate Human-Human-Object Interactions from Textual Descriptions (NeurIPS 2025 · similarity 0.85)
- HOIAnimator: Generating Text-prompt Human-object Animations using Novel Perceptive Diffusion Models (CVPR 2024 · similarity 0.85)
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.
