| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| text-to-motion generation | HumanML3D (test) | FID0 | 481 | |
| text-to-motion mapping | HumanML3D (test) | FID0.002 | 283 | |
| Motion Control | HumanML3D (test) | Average Error0 | 65 | |
| Text-to-motion generation | HumanML3D | R-Precision (Top 1)58.1 | 64 | |
| Text-driven Motion Generation | HumanML3D (test) | R-Precision@158.1 | 54 | |
| Text-to-Motion Synthesis | HumanML3D | R-Precision (Top 1)78.6 | 43 | |
| Motion Sequence Interpolation | HumanML3D | PSNR36.982 | 40 | |
| Motion-to-Text | HumanML3D (test) | BLEU@4100 | 40 | |
| Motion completion | HumanML3D (test) | MPJPE0 | 40 | |
| Text-to-Motion Generation | HumanML3D 19 (test) | FID0.0016 | 37 | |
| Motion-to-text retrieval | HumanML3D (test) | R@176.73 | 33 | |
| text-to-motion generation | HumanML3D 1 (test) | R-Precision (Top 1)0.517 | 32 | |
| Text-to-motion retrieval | HumanML3D (test) | R@178.21 | 30 | |
| Motion Description | HumanML3D (test) | BLEU-169.9 | 27 | |
| Motion-to-text retrieval | HumanML3D 1.0 (test) | R@168.64 | 24 | |
| Text-to-motion retrieval | HumanML3D 1.0 (test) | R@168.58 | 24 | |
| Text-to-motion generation | HumanML3D full dimension (test) | R-Precision Top 152.3 | 20 | |
| Text-to-Motion Generation (Kinematic Representation) | HumanML3D Kinematic Representation (test) | R-Precision@10.603 | 19 | |
| Text-to-Motion Generation | HumanML3D (Retain Set) | FID0.064 | 17 | |
| Text-to-Motion Generation | HumanML3D (Forget Set) | FID0.44 | 17 | |
| Text-to-motion generation | HumanML3D MARDM-67 evaluator (test) | FID0 | 16 | |
| Text-conditional Motion Synthesis | HumanML3D 16 (test) | R-Precision Top-10.525 | 15 | |
| Text-conditional motion synthesis | HumanML3D 12 (test) | R-Precision Top-151.5 | 15 | |
| Text-to-Motion Retrieval | HumanML3D | Recall@384.1 | 14 | |
| Text-to-motion | HumanML3D 10 (test) | R-Precision@177.8 | 12 |