Research map / 3D vision / 3D generation
Diffusion Implicit Policy for Unpaired Scene-aware Motion Synthesis
Jingyu Gong, Chong Zhang, Fengqi Liu, Ke Fan, Qianyu Zhou, Xin Tan, Zhizhong Zhang, Yuan Xie
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 |
|---|---|---|---|
| use DIP instead of DIMOS for locomotion finish time | DIP · Locomotion (ShapeNet scenes) | finish time (s): 6.43 -> 3.35 | Improved |
| DIP versus DIMOS on foot contact score | DIP · Locomotion (ShapeNet scenes) | foot contact score: 0.99 -> 0.91 | Hurt |
| DIP versus DIMOS on sitting max penetration | DIP · Sitting interaction (ShapeNet) | max penetration: 10.61 -> 7.13 | Improved |
| DIP versus DIMOS on lying max penetration | DIP · Lying interaction (ShapeNet) | max penetration: 44.61 -> 30.8 | Improved |
| DIP versus DIMOS on user-rated naturalness | DIP · Long-term synthesis user study (PROX/Replica) | naturalness rating: 2.72/3.09 -> 2.81/3.03 | No clear effect |
| DIP versus DIMOS on user-rated interaction plausibility | DIP · Long-term synthesis user study (PROX/Replica) | plausibility rating: 2.55/2.85 -> 3.15/3.17 | Improved |
| optimize mean via x0 prediction (GAN-inversion style) instead of directly modifying the mean (ablation) | DIP · scene-aware motion synthesis | motion quality | Improved |
Closest papers by meaning
- Enhanced Fine-Grained Motion Diffusion for Text-Driven Human Motion Synthesis (AAAI 2024 · similarity 0.87)
- ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models (CVPR 2024 · similarity 0.87)
- Optimizing Diffusion Noise Can Serve As Universal Motion Priors (CVPR 2024 · similarity 0.87)
- Move Anything with Layered Scene Diffusion (CVPR 2024 · similarity 0.87)
- Human Motion Diffusion as a Generative Prior (ICLR 2024 · similarity 0.87)
- Towards Detailed Text-to-Motion Synthesis via Basic-to-Advanced Hierarchical Diffusion Model (AAAI 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.
