3D Human Pose Estimation on HumanEva-I (test)
8.7Walking S1 Error (mm)GLA-GCN
Evaluation results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GLA-GCNProtocol=#2, Temporal information=true, Number of frames (T)=27, Input 2D pose (Detector)=GT2023.07 | 8.7 | 6.8 | 11.5 | 10.1 | 8.2 | 9.9 | 9.2 | — | — | — | — | — | — | — | — | — | — | |
| Li et al.Protocol=#2, Temporal information=true, Intermediate 3D pose sequence reconstruction=true, Input 2D pose (Detector)=GT2023.07 | 9.7 | 7.6 | 15.8 | 12.3 | 9.4 | 11.2 | 11.1 | — | — | — | — | — | — | — | — | — | — | |
| GLA-GCNProtocol=#2, Temporal information=true, Number of frames (T)=27, Input 2D pose (Detector)=MRCNN2023.07 | 12.5 | 9.1 | 26.9 | 18.5 | 12.7 | 12.8 | 15.4 | — | — | — | — | — | — | — | — | — | — | |
| Zhang et al.Protocol=#2, Temporal information=true, Intermediate 3D pose sequence reconstruction=true2023.07 | 12.7 | 10.9 | 17.6 | 22.6 | 15.8 | 17 | 16.1 | — | — | — | — | — | — | — | — | — | — | |
| Liu et al.Protocol=#2, Temporal information=true2023.07 | 13.1 | 9.8 | 26.8 | 16.9 | 12.8 | 13.3 | 15.5 | — | — | — | — | — | — | — | — | — | — | |
| VideoPose3D (MA)Protocol=Protocol 2, Training regime=Multi-action (MA), 2D detections=Mask R-CNN2018.11 | 13.9 | 10.2 | 46.6 | 20.9 | 13.1 | 13.8 | — | 23.8 | 33.7 | 32 | — | — | — | — | — | — | — | |
| Pavllo et al.Protocol=#2, Temporal information=true2023.07 | 13.9 | 10.2 | 46.6 | 20.9 | 13.1 | 13.8 | 19.8 | — | — | — | — | — | — | — | — | — | — | |
| Li et al.Protocol=#2, Temporal information=true, Intermediate 3D pose sequence reconstruction=true2023.07 | 14 | 10 | 32.8 | 19.5 | 13.6 | 14.2 | 17.4 | — | — | — | — | — | — | — | — | — | — | |
| Zheng et al.Protocol=#2, Temporal information=true2023.07 | 14.4 | 10.2 | 46.6 | 22.7 | 13.4 | 13.4 | 20.1 | — | — | — | — | — | — | — | — | — | — | |
| VideoPose3D (SA)Protocol=Protocol 2, Training regime=Single-action (SA), 2D detections=Mask R-CNN2018.11 | 14.5 | 10.5 | 47.3 | 21.9 | 13.4 | 13.9 | — | 24.3 | 34.9 | 32.1 | — | — | — | — | — | — | — | |
| Lee et al.Protocol=Protocol 2, Training regime=Multi-action (MA)2018.11 | 18.6 | 19.9 | 30.5 | 25.7 | 16.8 | 17.7 | — | 42.8 | 48.1 | 53.4 | — | — | — | — | — | — | — | |
| Lee et al.Protocol=#2, Temporal information=true2023.07 | 18.6 | 19.9 | 30.5 | 25.7 | 16.8 | 17.7 | 21.5 | — | — | — | — | — | — | — | — | — | — | |
| Ordinal Depth Supervisiontraining_data=HumanEva-I and LSP+MPII Ordinal images2018.05 | 18.8 | 12.7 | 29.2 | 23.5 | 15.4 | 14.5 | 18.3 | — | — | — | — | — | — | — | — | — | — | |
| ordinal depth supervisiontraining_strategy=mixed training (HumanEva-I + LSP+MPII Ordinal)2018.05 | 18.8 | 12.7 | 29.2 | 23.5 | 15.4 | 14.5 | 18.3 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al. (+)Protocol=Protocol 2, Training regime=Multi-action (MA), Extra data=true2018.11 | 18.8 | 12.7 | 29.2 | 23.5 | 15.4 | 14.5 | — | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.Protocol=#22023.07 | 18.8 | 12.7 | 29.2 | 23.5 | 15.4 | 14.5 | 19 | — | — | — | — | — | — | — | — | — | — | |
| sequence-to-sequence network2017.11 | 19.1 | 13.6 | 43.9 | 23.2 | 16.9 | 15.5 | 22 | — | — | — | — | — | — | — | — | — | — | |
| Pose Grammar2017.10 | 19.4 | 16.8 | 37.4 | 30.4 | 17.6 | 16.3 | 22.9 | — | — | — | — | — | — | — | — | — | — | |
| Fang et al.2017.11 | 19.4 | 16.8 | 37.4 | 30.4 | 17.6 | 16.3 | 22.9 | — | — | — | — | — | — | — | — | — | — | |
| Fang et al.Protocol=#22023.07 | 19.4 | 16.8 | 37.4 | 30.4 | 17.6 | 16.3 | 23 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2018.05 | 19.7 | 13 | 24.9 | 39.7 | 20 | 21 | 26.9 | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.2018.05 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2018.05 | 19.7 | 13 | 24.9 | 39.7 | 20 | 21 | 26.9 | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.2018.05 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2017.05 | 19.7 | 13 | 24.9 | 39.7 | 20 | 21 | 26.9 | — | — | — | — | — | — | — | — | — | — | |
| 3d-pose-baseline2D joint detector=Stacked Hourglass2017.05 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2017.05 | 19.7 | 13.5 | 26.5 | 34.6 | 17.9 | 20.1 | 22 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2017.10 | 19.7 | 13 | 24.9 | 39.7 | 20 | 21 | 26.9 | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.2017.10 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Moreno-Noguer2017.11 | 19.7 | 13 | 24.9 | 39.7 | 20 | 21 | 26.9 | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.2017.11 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.Protocol=Protocol 2, Training regime=Single-action (SA)2018.11 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | — | — | — | — | — | — | — | — | — | — | — | |
| Martinez et al.Protocol=#22023.07 | 19.7 | 17.4 | 46.8 | 26.9 | 18.2 | 18.6 | 24.6 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.2018.05 | 22.1 | 21.9 | 29 | 29.8 | 23.6 | 26 | 25.5 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.2018.05 | 22.1 | 21.9 | 29 | 29.8 | 23.6 | 26 | 25.5 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.2017.05 | 22.1 | 21.9 | 29 | 29.8 | 23.6 | 26 | 25.5 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.2017.11 | 22.1 | 21.9 | 29 | 29.8 | 23.6 | 26 | 25.5 | — | — | — | — | — | — | — | — | — | — | |
| Coarse-to-Fine Volumetric Prediction2016.11 | 22.3 | 19.5 | 29.7 | 28.9 | 21.9 | 23.8 | 24.3 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.2017.10 | 22.3 | 19.5 | 29.7 | 28.9 | 21.9 | 23.8 | 24.3 | — | — | — | — | — | — | — | — | — | — | |
| Pavlakos et al.Protocol=Protocol 2, Training regime=Multi-action (MA)2018.11 | 22.3 | 19.5 | 29.7 | 28.9 | 21.9 | 23.8 | — | — | — | — | — | — | — | — | — | — | — | |
| Lin et al.2017.11 | 26.5 | 20.7 | 38 | 41 | 29.7 | 29.1 | 30.8 | — | — | — | — | — | — | — | — | — | — | |
| Popa et al.2017.05 | 27.1 | 18.4 | 39.5 | 37.6 | 28.9 | 27.6 | 29.9 | — | — | — | — | — | — | — | — | — | — | |
| Dual-Source Approach2017.05 | 27.4 | 28.6 | 32.5 | 39.9 | 29.4 | 31.4 | 31.5 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2016.11 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2018.05 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2018.05 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2017.05 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2017.05 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2017.10 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.2017.11 | 35.8 | 32.4 | 41.6 | 46.6 | 41.4 | 35.4 | 38.9 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.Evaluation protocol=different (*)2017.05 | 38.2 | 32.8 | 40.2 | 42 | 34.7 | 46.4 | 39.1 | — | — | — | — | — | — | — | — | — | — | |
| Dual-Source ApproachTraining Data=MoCap from CMU dataset2017.05 | 39.5 | 47.3 | 61.4 | 55.5 | 48.3 | 53.1 | 50.5 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2016.11 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2018.05 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2018.05 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2017.05 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2017.05 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2017.10 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Kostrikov et al.2017.11 | 44 | 30.9 | 41.7 | 57.2 | 35 | 33.3 | 40.3 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.2016.11 | 46.4 | 30.3 | 64.9 | 64.5 | 48 | 38.2 | 48.7 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.2018.05 | 46.4 | 30.3 | 64.9 | 64.5 | 48 | 38.2 | 48.7 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.2018.05 | 46.4 | 30.3 | 64.9 | 64.5 | 48 | 38.2 | 48.7 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.2017.05 | 46.4 | 30.3 | 64.9 | 64.5 | 48 | 38.2 | 48.7 | — | — | — | — | — | — | — | — | — | — | |
| Bo et al.2017.11 | 46.4 | 30.3 | 64.9 | 64.5 | 48 | 38.2 | 48.7 | — | — | — | — | — | — | — | — | — | — | |
| Yasin et al.Training Data=MoCap from CMU dataset2017.05 | 52.2 | 51 | 62.8 | 74.5 | 72.4 | 56.8 | 61.6 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2016.11 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2018.05 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2018.05 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2017.05 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.citation=[49]2017.05 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2017.10 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.2017.11 | 65.1 | 48.6 | 73.5 | 74.2 | 46.6 | 32.2 | 56.7 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2016.11 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2018.05 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2018.05 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2017.05 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2017.05 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Wang et al.2017.11 | 71.9 | 75.7 | 85.3 | 62.6 | 77.7 | 54.4 | 71.3 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2016.11 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2018.05 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2018.05 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2017.05 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2017.05 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Radwan et al.2017.11 | 75.1 | 99.8 | 93.8 | 79.2 | 89.8 | 99.4 | 89.5 | — | — | — | — | — | — | — | — | — | — | |
| Simo-Serra et al.citation=[50]2017.05 | 99.6 | 108.3 | 127.4 | 109.2 | 93.1 | 115.8 | 108.9 | — | — | — | — | — | — | — | — | — | — | |
| HUND+SO+GTModel Variant=Baseline2022.05 | — | — | — | — | — | — | 208 | — | — | — | 90 | 76 | 14 | 0.51 | 40 | — | — | |
| HUND+SO+GT + DynamicsModel Variant=Full Model2022.05 | — | — | — | — | — | — | 196 | — | — | — | 91 | 74 | 14 | 0.27 | 4 | — | — | |
| PoseFormerf (number of frames)=92022.03 | — | — | — | — | — | — | — | — | — | — | 104.7 | — | — | — | — | 1,470.4 | 1,452 | |
| PoseFormerf=92022.03 | — | — | — | — | — | — | — | — | — | — | 79.4 | — | — | — | — | 260.7 | 253.9 | |
| PoseLifterf (number of frames)=12022.03 | — | — | — | — | — | — | — | — | — | — | 240.2 | — | — | — | — | 1,263 | 1,129.3 | |
| PoseLifterf=12022.03 | — | — | — | — | — | — | — | — | — | — | 3,690.2 | — | — | — | — | 15,170.6 | 16,082.6 | |
| Ray3Df (number of frames)=92022.03 | — | — | — | — | — | — | — | — | — | — | 83.9 | — | — | — | — | 477.8 | 468.6 | |
| Ray3Df=92022.03 | — | — | — | — | — | — | — | — | — | — | 81.5 | — | — | — | — | 121.5 | 99.3 | |
| Rempe et al.Model Variant=Kinematics2022.05 | — | — | — | — | — | — | 408 | — | — | — | — | — | — | — | — | — | — | |
| Rempe et al.Model Variant=Dynamics2022.05 | — | — | — | — | — | — | 422 | — | — | — | — | — | — | — | — | — | — | |
| RIEf (number of frames)=92022.03 | — | — | — | — | — | — | — | — | — | — | 92 | — | — | — | — | 1,375.8 | 1,369.5 | |
| RIEf=92022.03 | — | — | — | — | — | — | — | — | — | — | 82 | — | — | — | — | 272.9 | 285.1 | |
| Videoposef (number of frames)=92022.03 | — | — | — | — | — | — | — | — | — | — | 86.2 | — | — | — | — | 1,387.4 | 1,387.1 | |
| Videoposef=92022.03 | — | — | — | — | — | — | — | — | — | — | 85.1 | — | — | — | — | 284.6 | 283.1 |