Motion Synthesis On Humanml3D
评估指标
Diversity
FID
Multimodality
R Precision Top3
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Diversity | FID | Multimodality | R Precision Top3 |
|---|---|---|---|---|
| human-motion-diffusion-model | 9.559 | 0.544 | 2.799 | 0.611 |
| finemogen-fine-grained-spatio-temporal-motion-1 | 9.263 | 0.151 | 2.696 | 0.784 |
| motionlcm-real-time-controllable-motion | 9.607 | 0.304 | 2.259 | 0.798 |
| act-as-you-wish-fine-grained-control-of | 9.692 | 0.116 | 2.766 | 0.785 |
| motiondiffuse-text-driven-human-motion | 9.410 | 0.630 | 1.553 | 0.782 |
| attt2m-text-driven-human-motion-generation-1 | 9.700 | 0.112 | 2.452 | 0.786 |
| t2m-gpt-generating-human-motion-from-textual | 9.844 | 0.140 | 3.285 | 0.685 |
| executing-your-commands-via-motion-diffusion | 9.724 | 0.473 | 2.413 | 0.772 |
| remodiffuse-retrieval-augmented-motion | 9.018 | 0.103 | 1.795 | 0.795 |
| guess-gradually-enriching-synthesis-for-text | 9.826 | 0.109 | 2.430 | 0.787 |
| parco-part-coordinating-text-to-motion | 9.576 | 0.109 | 1.382 | 0.801 |
| motionpcm-real-time-motion-synthesis-with | 9.575 | 0.030 | 1.714 | 0.842 |
| tm2t-stochastic-and-tokenized-modeling-for | 8.589 | 1.501 | 2.424 | 0.729 |
| fg-t2m-fine-grained-text-driven-human-motion | 9.278 | 0.243 | 1.614 | 0.783 |
| mcm-multi-condition-motion-synthesis-1 | 9.585 | 0.053 | 0.8104 | 0.788 |
| emdm-efficient-motion-diffusion-model-for | 9.551 | 0.112 | 1.641 | 0.786 |
| generating-diverse-and-natural-3d-human | 9.175 | 1.087 | 2.219 | 0.736 |
| bad-bidirectional-auto-regressive-diffusion | 9.688 | 0.049 | 1.119 | 0.800 |
| bamm-bidirectional-autoregressive-motion | 9.717 | 0.055 | 1.687 | 0.814 |
| mmm-generative-masked-motion-model | 9.577 | 0.089 | 1.226 | 0.804 |
| diffusion-motion-generate-text-guided-3d | 23.692 | 10.21 | - | 0.735 |
| mmm-generative-masked-motion-model | 9.411 | 0.080 | 1.164 | 0.794 |
| t2m-gpt-generating-human-motion-from-textual | 9.761 | 0.116 | 1.856 | 0.775 |
| bipo-bidirectional-partial-occlusion-network-1 | 9.556 | 0.030 | 1.374 | 0.809 |
| tm2t-stochastic-and-tokenized-modeling-for | 6.409 | 5.012 | - | 0.345 |
| tm2t-stochastic-and-tokenized-modeling-for | 7.676 | 11.02 | - | 0.486 |
| momask-generative-masked-modeling-of-3d-human | - | 0.045 | 1.241 | 0.807 |
| motion-mamba-efficient-and-long-sequence | 9.871 | 0.281 | 2.294 | 0.792 |
| t2m-gpt-generating-human-motion-from-textual | 9.722 | 0.141 | 1.831 | 0.775 |
| make-an-animation-large-scale-text | 8.23 | 0.774 | - | 0.676 |
| bad-bidirectional-auto-regressive-diffusion | 9.694 | 0.065 | 1.194 | 0.808 |
| free-t2m-frequency-enhanced-text-to-motion | 9.480 | 0.051 | - | 0.803 |
| diversemotion-towards-diverse-human-motion | 9.551 | 0.070 | 2.062 | 0.783 |
| diversemotion-towards-diverse-human-motion | 9.683 | 0.072 | 1.869 | 0.802 |
| motion-anything-any-to-motion-generation | 9.521 | 0.028 | 2.705 | 0.829 |
| discord-discrete-tokens-to-continuous-motion | - | 0.032 | 1.288 | 0.809 |
| tm2d-bimodality-driven-3d-dance-generation | 9.513 | 1.021 | 4.139 | - |