Motion Synthesis On Kit Motion Language
평가 지표
Diversity
FID
Multimodality
R Precision Top3
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Diversity | FID | Multimodality | R Precision Top3 |
|---|---|---|---|---|
| parco-part-coordinating-text-to-motion | 10.95 | 0.453 | 1.245 | 0.772 |
| bad-bidirectional-auto-regressive-diffusion | 11.000 | 0.221 | 1.170 | 0.750 |
| mmm-generative-masked-motion-model | 10.910 | 0.316 | 1.232 | 0.744 |
| executing-your-commands-via-motion-diffusion | 10.84 | 3.717 | 0.532 | 0.687 |
| generating-diverse-and-natural-3d-human | 10.72 | 3.022 | 2.052 | 0.681 |
| discord-discrete-tokens-to-continuous-motion | - | 0.169 | 1.266 | 0.775 |
| motionpcm-real-time-motion-synthesis-with | 10.827 | 0.294 | 1.254 | 0.787 |
| motion-anything-any-to-motion-generation | 10.94 | 0.131 | 1.374 | 0.802 |
| remodiffuse-retrieval-augmented-motion | 10.80 | 0.155 | 1.239 | 0.765 |
| fg-t2m-fine-grained-text-driven-human-motion | 10.93 | 0.571 | 1.019 | 0.745 |
| tm2t-stochastic-and-tokenized-modeling-for | 9.334 | 12.12 | - | 0.338 |
| mmm-generative-masked-motion-model | 10.633 | 0.429 | 1.105 | 0.718 |
| finemogen-fine-grained-spatio-temporal-motion-1 | 10.85 | 0.178 | 1.877 | 0.772 |
| momask-generative-masked-modeling-of-3d-human | - | 0.204 | 1.131 | 0.781 |
| free-t2m-frequency-enhanced-text-to-motion | 10.902 | 0.155 | - | 0.789 |
| bipo-bidirectional-partial-occlusion-network-1 | 10.833 | 0.164 | 1.098 | 0.803 |
| bamm-bidirectional-autoregressive-motion | 11.008 | 0.183 | 1.609 | 0.788 |
| diversemotion-towards-diverse-human-motion | 10.873 | 0.468 | 2.062 | 0.760 |
| motiondiffuse-text-driven-human-motion | 11.10 | 1.954 | 0.730 | 0.739 |
| emdm-efficient-motion-diffusion-model-for | 10.96 | 0.261 | 1.343 | 0.780 |
| human-motion-diffusion-model | 10.847 | 0.497 | 1.907 | 0.396 |
| t2m-gpt-generating-human-motion-from-textual | 10.862 | 0.717 | 1.912 | 0.737 |
| tm2t-stochastic-and-tokenized-modeling-for | 9.473 | 3.599 | 3.292 | 0.587 |
| t2m-gpt-generating-human-motion-from-textual | 11.198 | 0.737 | 2.309 | 0.716 |
| guess-gradually-enriching-synthesis-for-text | 10.933 | 0.371 | 2.732 | 0.751 |
| tm2t-stochastic-and-tokenized-modeling-for | 9.073 | 6.545 | - | 0.483 |
| attt2m-text-driven-human-motion-generation-1 | 10.96 | 0.870 | 2.281 | 0.751 |
| t2m-gpt-generating-human-motion-from-textual | 10.921 | 0.514 | 1.570 | 0.745 |
| executing-your-commands-via-motion-diffusion | 10.80 | 0.404 | 2.192 | 0.734 |
| act-as-you-wish-fine-grained-control-of | 11.12 | 0.313 | 3.627 | 0.769 |