Motion Synthesis On Interhuman
المقاييس
FID
MMDist
MModality
R-Precision Top3
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
اسم النموذج | FID | MMDist | MModality | R-Precision Top3 | Paper Title | Repository |
|---|---|---|---|---|---|---|
| in2IN | 5.177 | 3.790 | 1.061 | 0.687 | in2IN: Leveraging individual Information to Generate Human INteractions | - |
| MDM | 9.167 | 7.125 | 2.35 | 0.339 | Human Motion Diffusion Model | |
| ComMDM | 7.069 | 6.212 | 1.822 | 0.466 | Human Motion Diffusion as a Generative Prior | |
| FreeMotion | 6.740 | 3.848 | 1.226 | 0.544 | FreeMotion: A Unified Framework for Number-free Text-to-Motion Synthesis | - |
| T2M | 13.769 | 5.731 | 1.387 | 0.464 | Generating Diverse and Natural 3D Human Motions From Text | |
| TEMOS | 17.375 | 6.342 | 0.535 | 0.450 | TEMOS: Generating diverse human motions from textual descriptions | |
| InterMask | 5.154 | 3.790 | 1.737 | 0.683 | InterMask: 3D Human Interaction Generation via Collaborative Masked Modelling | - |
| InterGen | 5.918 | 5.108 | 2.141 | 0.624 | InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactions | |
| MoMat-MoGen | 5.674 | 3.790 | 1.295 | 0.666 | Digital Life Project: Autonomous 3D Characters with Social Intelligence | - |
| RIG | 6.775 | 5.876 | 2.141 | 0.521 | Role-Aware Interaction Generation from Textual Description |
0 of 10 row(s) selected.