3D Pose Estimation on 3DHP
75.3MPJPELifelong Domain Adaptive 3D HPE framework
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Lifelong Domain Adaptive 3D HPE frameworkTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 75.3 | — | — | — | — | 51.1 | |
| Lifelong Domain Adaptive 3D HPE frameworkTime=t = 62025.12 | 75.3 | — | — | — | — | 50.7 | |
| Lifelong Domain Adaptive 3D HPE frameworkTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 78.3 | — | — | — | — | 52.2 | |
| PoseDA-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 79.8 | — | — | — | — | 53 | |
| RMT-PoseTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 79.9 | — | — | — | — | 53.3 | |
| CycleAdapt-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 80 | — | — | — | — | 53.3 | |
| Adaptpose-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 80.5 | — | — | — | — | 53.4 | |
| PoseDA-LLTime=t = 62025.12 | 80.7 | — | — | — | — | 54.5 | |
| CoTTA-PoseTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 81.1 | — | — | — | — | 53.6 | |
| CoTTA-PoseTime=t = 62025.12 | 81.3 | — | — | — | — | 55.2 | |
| CycleAdapt-LLTime=t = 62025.12 | 81.4 | — | — | — | — | 55.4 | |
| RMT-PoseTime=t = 62025.12 | 81.6 | — | — | — | — | 55.5 | |
| Adaptpose-LLTime=t = 62025.12 | 82.2 | — | — | — | — | 56.1 | |
| PoseDA-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 88.9 | — | — | — | — | 62.1 | |
| CoTTA-PoseTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 90 | — | — | — | — | 63.8 | |
| CycleAdapt-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 90.2 | — | — | — | — | 64.1 | |
| Adaptpose-LLTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 90.5 | — | — | — | — | 64.1 | |
| RMT-PoseTime (t)=2, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 90.9 | — | — | — | — | 64.3 | |
| Source-onlyTime (t)=0, Adaptation Sequence=H3.6M (Source) → 3DHP → 3DPW2025.12 | 96.4 | — | — | — | — | 66.5 | |
| Source-onlyTime (t)=0, Adaptation Sequence=H3.6M (Source) → 3DPW → 3DHP2025.12 | 96.4 | — | — | — | — | 66.5 | |
| Semantics-aware Test-time AdaptationBackbone=HMR, Pre-training Dataset=Human3.6M2025.02 | 101.3 | — | — | — | — | 65.1 | |
| CycleAdaptBackbone=HMR, Pre-training Dataset=Human3.6M2025.02 | 110.3 | — | — | — | — | 74.4 | |
| DynaBOABackbone=HMR, Pre-training Dataset=Human3.6M2025.02 | 130.7 | — | — | — | — | 81.8 | |
| BOABackbone=HMR, Pre-training Dataset=Human3.6M2025.02 | 135.3 | — | — | — | — | 88.5 | |
| Pre-trained HMRBackbone=HMR, Pre-training Dataset=Human3.6M2025.02 | 218.5 | — | — | — | — | 119.6 | |
| GFPoseNumber of samples=2002022.12 | — | 88.4 | 87.1 | 84.3 | 86.9 | — | |
| Li et al. [28]Number of samples=2002022.12 | — | 70.1 | 68.2 | 66.6 | 67.9 | — | |
| Li et al. [29]Number of samples=2002022.12 | — | 86.9 | 86.6 | 79.3 | 85 | — | |
| Wehrbein et al. [60]Number of samples=2002022.12 | — | 86.6 | 82.8 | 82.5 | 84.3 | — |