Research map / 3D vision / 3D generation
EigenActor: Variant Body-Object Interaction Generation Evolved from Invariant Action Basis Reasoning
Xuehao Gao, Yang Yang, Shaoyi Du, Yang Wu, Yebin Liu, Guo-Jun Qi
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 action-specific motion basis inference (ablation) | EigenActor · GRAB | R-Precision (Top-3): 0.48 -> 0.66 | Improved |
| add action-specific motion basis inference (ablation) | EigenActor · GRAB | FID: 2.31 -> 0.89 | Improved |
| use real instead of inferred motion basis at test (ablation) | EigenActor · GRAB | FID: 0.89 -> 0.82 | No clear effect |
| add object contact part inference (ablation) | EigenActor · GRAB | FID: 1.99 -> 0.89 | Improved |
| add object contact part inference (ablation) | EigenActor · GRAB | R-Precision (Top-3): 0.59 -> 0.66 | Improved |
| use joint text and object condition instead of text-only (ablation) | object-specific interaction diffusion · GRAB | R-Precision (Top-3): 0.58 -> 0.66 | Improved |
| use joint text and object condition instead of object-only (ablation) | object-specific interaction diffusion · GRAB | R-Precision (Top-3): 0.61 -> 0.66 | Improved |
| add temporal consistency constraint (ablation) | interaction optimization · GRAB | FID: 1.72 -> 0.93 | Improved |
| add in-contact constraint (ablation) | interaction optimization · GRAB | FID: 1.72 -> 0.95 | Improved |
| add both constraints (ablation) | interaction optimization · GRAB | contact precision: 0.49 -> 0.72 | Improved |
| increase denoiser layers from 4 to 8 (ablation) | EigenActor · GRAB | R-Precision (Top-3): 0.58 -> 0.66 | Improved |
| increase denoiser layers from 8 to 10 (ablation) | EigenActor · GRAB | R-Precision (Top-3): 0.66 -> 0.67 | No clear effect |
Closest papers by meaning
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.91)
- CG-HOI: Contact-Guided 3D Human-Object Interaction Generation (CVPR 2024 · similarity 0.90)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.90)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (CVPR 2025 · similarity 0.89)
- NIFTY: Neural Object Interaction Fields for Guided Human Motion Synthesis (CVPR 2024 · similarity 0.87)
- ChainHOI: Joint-based Kinematic Chain Modeling for Human-Object Interaction Generation (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.
