Text-to-motion mapping on KIT-ML (test)
86.34R Precision (Top 3)Ours(3-modal)
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ours(3-modal)Protocol=Small batches2024.03 | 86.34 | 58.1 | 77.8 | — | — | — | — | — | — | 93.08 | 96.47 | 1.08 | — | |
| Ours(2-modal)Protocol=Small batches2024.03 | 82.05 | 53.96 | 76.42 | — | — | — | — | — | — | 89.66 | 95.22 | 1.38 | — | |
| TMRProtocol=Small batches2024.03 | 78.25 | 49.25 | 69.75 | — | — | — | — | — | — | 87.88 | 95 | 1.5 | — | |
| Guo et al.Protocol=Small batches2024.03 | 75.12 | 42.25 | 62.62 | — | — | — | — | — | — | 87.5 | 96.12 | 1.88 | — | |
| MotionCLIPProtocol=Small batches2024.03 | 69.5 | 41.29 | 55.38 | — | — | — | — | — | — | 78.83 | 90.12 | 1.73 | — | |
| TEMOSProtocol=Small batches2024.03 | 67 | 43.88 | 58.25 | — | — | — | — | — | — | 74 | 84.75 | 2.06 | — | |
| Ours(3-modal)Protocol=Dissimilar subset2024.03 | 63 | 30 | 49 | — | — | — | — | — | — | 73 | 84 | 3 | — | |
| TMRProtocol=Dissimilar subset2024.03 | 60 | 26 | 46 | — | — | — | — | — | — | 70 | 83 | 3 | — | |
| Ours(2-modal)Protocol=Dissimilar subset2024.03 | 60 | 29 | 45 | — | — | — | — | — | — | 71 | 81 | 2 | — | |
| Ours(3-modal)Protocol=All with threshold2024.03 | 48.63 | 30.86 | 41.8 | — | — | — | — | — | — | 59.96 | 74.22 | 4 | — | |
| TEMOSProtocol=Dissimilar subset2024.03 | 46 | 24 | 40 | — | — | — | — | — | — | 54 | 70 | 5 | — | |
| Ours(2-modal)Protocol=All with threshold2024.03 | 42.73 | 24.02 | 30.86 | — | — | — | — | — | — | 54.69 | 70.09 | 5 | — | |
| TMRProtocol=All with threshold2024.03 | 41.93 | 24.58 | 30.24 | — | — | — | — | — | — | 50.48 | 60.36 | 5 | — | |
| MotionCLIPProtocol=Dissimilar subset2024.03 | 41 | 19 | 33 | — | — | — | — | — | — | 50 | 69 | 6 | — | |
| Guo et al.Protocol=Dissimilar subset2024.03 | 36 | 16 | 29 | — | — | — | — | — | — | 48 | 66 | 6 | — | |
| MotionCLIPProtocol=All with threshold2024.03 | 31.45 | 13.79 | 23.08 | — | — | — | — | — | — | 42.93 | 53.01 | 9 | — | |
| TEMOSProtocol=All with threshold2024.03 | 30.84 | 18.55 | 24.34 | — | — | — | — | — | — | 42.29 | 56.37 | 7 | — | |
| Guo et al.Protocol=All with threshold2024.03 | 29.76 | 13.25 | 22.65 | — | — | — | — | — | — | 39.04 | 49.52 | 11 | — | |
| Ours(3-modal)Protocol=All2024.03 | 24.61 | 10.16 | 19.92 | — | — | — | — | — | — | 34.57 | 49.8 | 11 | — | |
| Ours(2-modal)Protocol=All2024.03 | 21.09 | 8.59 | 15.04 | — | — | — | — | — | — | 32.23 | 46.09 | 13 | — | |
| TMRProtocol=All2024.03 | 20.36 | 7.23 | 13.98 | — | — | — | — | — | — | 28.31 | 40.12 | 17 | — | |
| TEMOSProtocol=All2024.03 | 17.59 | 7.11 | 13.25 | — | — | — | — | — | — | 24.1 | 35.66 | 24 | — | |
| MotionCLIPProtocol=All2024.03 | 14.36 | 4.87 | 9.31 | — | — | — | — | — | — | 20.09 | 31.57 | 26 | — | |
| Guo et al.Protocol=All2024.03 | 10.84 | 3.37 | 6.99 | — | — | — | — | — | — | 16.87 | 27.71 | 28 | — | |
| BiPOGround-truth motion length=false2024.11 | 0.803 | 0.444 | 0.674 | 0.164 | 2.658 | 10.833 | 1.098 | — | — | — | — | — | — | |
| Free-T2MVenue=-2025.01 | 0.798 | 0.454 | 0.675 | 0.15 | 2.689 | 10.977 | 1.141 | — | — | — | — | — | — | |
| BAMMGround-truth motion length=true2024.03 | 0.791 | 0.436 | 0.66 | 0.2 | 2.714 | 10.914 | 1.517 | — | — | — | — | — | — | |
| BAMM2024.03 | 0.788 | 0.438 | 0.661 | 0.183 | 2.723 | 11.008 | 1.609 | — | — | — | — | — | — | |
| BAMMGround-truth motion length=false2024.11 | 0.788 | 0.438 | 0.661 | 0.183 | 2.723 | 11.008 | 1.609 | — | — | — | — | — | — | |
| T2M Mamba2026.02 | 0.784 | 0.437 | 0.659 | 0.195 | 2.756 | 10.92 | 1.512 | — | — | — | — | — | — | |
| StableMoFusionVenue=ACM MM 20242025.01 | 0.782 | 0.445 | 0.66 | 0.258 | 2.8 | 10.836 | 1.362 | — | — | — | — | — | — | |
| MoMaskbase-layer tokens only=false2023.11 | 0.781 | 0.433 | 0.656 | 0.204 | 2.779 | — | — | 1.131 | — | — | — | — | — | |
| MoMaskGround-truth motion length=true2024.03 | 0.781 | 0.433 | 0.656 | 0.204 | 2.779 | 11.311 | 1.131 | — | — | — | — | — | — | |
| MoMaskVenue=CVPR 20242025.01 | 0.781 | 0.433 | 0.656 | 0.204 | 2.779 | — | 1.131 | — | — | — | — | — | — | |
| MoMaskGround-truth motion length=true2024.11 | 0.781 | 0.433 | 0.656 | 0.204 | 2.779 | 11.08 | 1.131 | — | — | — | — | — | — | |
| MoMask2026.02 | 0.781 | 0.433 | 0.656 | 0.204 | 2.779 | — | 1.131 | — | — | — | — | — | — | |
| Real motionsmotion length=ground truth2022.08 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real2022.09 | 0.779 | — | — | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motions2023.04 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real2022.12 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real2023.06 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motion2023.09 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motion2023.01 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motions2023.12 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motions2023.12 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motion2024.01 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motion2024.01 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real2023.12 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motion2024.03 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real2024.09 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Ground TruthVenue=-2025.01 | 0.779 | 0.424 | 0.649 | 0.031 | — | 11.08 | — | — | — | — | — | — | — | |
| Real motion2024.11 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| Real motions2022.07 | 0.779 | 0.424 | 0.649 | 0.031 | 2.788 | 11.08 | — | — | — | — | — | — | — | |
| FineMoGen2023.12 | 0.772 | 0.432 | 0.649 | 0.178 | 2.869 | 10.85 | — | 1.877 | — | — | — | — | — | |
| FineMoGen2023.12 | 0.772 | 0.432 | 0.649 | 0.178 | 2.869 | 10.85 | — | — | 1.877 | — | — | — | — | |
| ParCo2024.03 | 0.772 | 0.43 | 0.649 | 0.453 | 2.82 | 10.95 | 1.245 | — | — | — | — | — | — | |
| ParCoGround-truth motion length=false2024.11 | 0.772 | 0.43 | 0.649 | 0.453 | 2.82 | 10.95 | 1.245 | — | — | — | — | — | — | |
| GraphMotionTotal diffusion steps=1502023.11 | 0.769 | 0.429 | 0.648 | 0.313 | 3.076 | 11.12 | 3.627 | — | — | — | — | — | — | |
| ReMoDiffuse2023.04 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | 10.8 | 1.239 | — | — | — | — | — | — | |
| ReMoDiffuse2023.11 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | — | — | 1.239 | — | — | — | — | — | |
| ReMoDiffuse2023.12 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | 10.8 | — | 1.239 | — | — | — | — | — | |
| ReMoDiffuse2023.12 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | 10.8 | — | — | 1.239 | — | — | — | — | |
| ReMoDiffuse2024.03 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | 10.8 | 1.239 | — | — | — | — | — | — | |
| ReMoDiffuseVenue=ICCV 20232025.01 | 0.765 | 0.427 | 0.641 | 0.155 | 3.008 | 10.8 | 1.239 | — | — | — | — | — | — | |
| ReMoDiffuseGround-truth motion length=true2024.11 | 0.765 | 0.427 | 0.641 | 0.155 | 2.814 | 10.8 | 1.239 | — | — | — | — | — | — | |
| Motion Mamba2026.02 | 0.765 | 0.419 | 0.645 | 0.307 | 3.021 | 11.02 | 1.678 | — | — | — | — | — | — | |
| Ours2023.09 | 0.76 | 0.416 | 0.637 | 0.468 | 2.892 | 10.873 | 2.062 | — | — | — | — | — | — | |
| MoMaskbase-layer tokens only=true2023.11 | 0.76 | 0.415 | 0.634 | 0.372 | 2.931 | — | — | 1.097 | — | — | — | — | — | |
| GraphMotionTotal diffusion steps=502023.11 | 0.755 | 0.417 | 0.635 | 0.262 | 3.085 | 11.21 | 3.568 | — | — | — | — | — | — | |
| AttT2M2023.09 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| GUESS2024.01 | 0.751 | 0.425 | 0.632 | 0.371 | 2.421 | 10.933 | 2.732 | — | — | — | — | — | — | |
| GUESS2024.01 | 0.751 | 0.425 | 0.632 | 0.371 | 2.421 | 10.933 | 2.732 | — | — | — | — | — | — | |
| AttT2M2023.12 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| AttT2M2024.03 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| AttT2M2024.03 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| AttT2M2024.09 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| AttT2MGround-truth motion length=false2024.11 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| AttT2M2026.02 | 0.751 | 0.413 | 0.632 | 0.87 | 3.039 | 10.96 | 2.281 | — | — | — | — | — | — | |
| BADSampling strategy=OAAS 2.3.12024.09 | 0.75 | 0.417 | 0.631 | 0.221 | 2.941 | 11 | 1.17 | — | — | — | — | — | — | |
| VQ-VAE2024.09 | 0.746 | 0.4 | 0.619 | 0.437 | 2.981 | 11.093 | — | — | — | — | — | — | — | |
| T2M-GPT2023.04 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| T2M-GPT2023.06 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | 1.57 | — | — | — | — | — | — | |
| VQ-VAE2023.09 | 0.745 | 0.402 | 0.614 | 0.374 | 2.942 | 10.955 | — | — | — | — | — | — | — | |
| T2M-GPT2023.09 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| T2M-GPT2023.09 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | 1.57 | — | — | — | — | — | — | |
| Fg-T2M2023.09 | 0.745 | 0.418 | 0.626 | 0.571 | 3.114 | 10.93 | 1.019 | — | — | — | — | — | — | |
| T2M-GPTtau=U[0, 1]2023.01 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| T2M-GPT2023.11 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | — | — | 1.57 | — | — | — | — | — | |
| T2M-GPT2023.11 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | 1.57 | — | — | — | — | — | — | |
| T2M-GPT2023.12 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | — | 1.57 | — | — | — | — | — | |
| T2M-GPT2023.12 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | — | — | 1.57 | — | — | — | — | |
| T2M-GPT2024.01 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| T2M-GPT2024.01 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| Fg-T2Mground-truth motion length=true2023.12 | 0.745 | 0.418 | 0.626 | 0.571 | 3.114 | 10.93 | 1.019 | — | — | — | — | — | — | |
| Fg-T2M2024.03 | 0.745 | 0.418 | 0.626 | 0.571 | 3.114 | 10.93 | 1.019 | — | — | — | — | — | — | |
| T2M-GPT2024.03 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | 1.57 | — | — | — | — | — | — | |
| Fg-T2M2024.03 | 0.745 | 0.418 | 0.626 | 0.571 | 3.114 | 10.93 | 1.019 | — | — | — | — | — | — | |
| T2M-GPTVenue=CVPR 20232025.01 | 0.745 | 0.416 | 0.627 | 0.141 | 3.007 | 10.921 | 1.57 | — | — | — | — | — | — | |
| Fg-T2MGround-truth motion length=true2024.11 | 0.745 | 0.418 | 0.626 | 0.571 | 3.114 | 10.93 | 1.019 | — | — | — | — | — | — | |
| T2M-GPT2026.02 | 0.745 | 0.416 | 0.627 | 0.514 | 3.007 | 10.92 | 1.57 | — | — | — | — | — | — |