Action Quality Assessment on AQA-7 (test)
89.23DivingDAE-CoRe
Evaluation results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DAE-CoRe2021.11 | 89.23 | 0.7786 | 0.7102 | 0.6842 | 0.9506 | 0.9129 | 0.852 | — | — | — | — | — | — | — | |
| CoRe2021.11 | 88.24 | 0.7746 | 0.7115 | 0.6624 | 0.9442 | 0.9078 | 0.8401 | — | — | — | — | — | — | — | |
| CoRe + GART*Use of Degree of Difficulty (DD)=false2021.08 | 88.24 | 77.46 | 0.7115 | 0.6624 | 0.9442 | 0.9078 | 0.8401 | 0.64 | 1.78 | 3.67 | 3.87 | 0.41 | 2.35 | 2.12 | |
| I3D + MLP*Implementation Source=Our Implementation (#), Use of Degree of Difficulty (DD)=false2021.08 | 86.85 | 69.39 | 0.5391 | 0.518 | 0.8782 | 0.8486 | 0.7601 | 0.81 | 2.54 | 6.06 | 5.31 | 1.41 | 3.08 | 3.2 | |
| DAE-MLP2021.11 | 84.2 | 0.7754 | 0.6836 | 0.723 | 0.9237 | 0.8902 | 0.8258 | — | — | — | — | — | — | — | |
| Ours-USDLapproach=Uncertainty-aware Soft Label Distribution Learning2020.06 | 80.99 | 75.7 | 65.38 | 71.09 | 91.66 | 88.78 | 0.8102 | — | — | — | — | — | — | — | |
| USDL2021.11 | 80.99 | 0.757 | 0.6538 | 0.7109 | 0.9166 | 0.8878 | 0.8102 | — | — | — | — | — | — | — | |
| USDLYear=20202021.08 | 80.99 | 75.7 | 0.6538 | 0.7109 | 0.9166 | 0.8878 | 0.8102 | 0.79 | 2.09 | 4.82 | 4.94 | 0.65 | 2.14 | 2.57 | |
| C3D-SVRTraining Protocol=Single-action, Aggregation Module=SVR, Feature Extraction Backbone=C3D2018.12 | 79.02 | 68.24 | 52.09 | 40.06 | 59.37 | 91.2 | 0.6937 | — | — | — | — | — | — | — | |
| C3D-SVR2020.06 | 79.02 | 68.24 | 52.09 | 40.06 | 59.37 | 91.2 | 0.6937 | — | — | — | — | — | — | — | |
| C3D-SVR2021.11 | 79.02 | 0.6824 | 0.5209 | 0.4006 | 0.5937 | 0.912 | 0.6937 | — | — | — | — | — | — | — | |
| C3D-SVRYear=20172021.08 | 79.02 | 68.24 | 0.5209 | 0.4006 | 0.5937 | 0.912 | 0.6937 | 1.53 | 3.12 | 6.79 | 7.03 | 17.84 | 4.83 | 6.86 | |
| JRG2020.06 | 76.3 | 73.58 | 60.06 | 54.05 | 90.13 | 92.54 | 0.7849 | — | — | — | — | — | — | — | |
| JRG2021.11 | 76.3 | 0.7358 | 0.6006 | 0.5405 | 0.9013 | 0.9254 | 0.7849 | — | — | — | — | — | — | — | |
| JRGYear=20192021.08 | 76.3 | 73.58 | 0.6006 | 0.5405 | 0.9013 | 0.9254 | 0.7849 | — | — | — | — | — | — | — | |
| Ours-Regressionapproach=Regression2020.06 | 74.38 | 73.42 | 51.9 | 51.03 | 89.15 | 87.03 | 0.7472 | — | — | — | — | — | — | — | |
| I3D+MLP*Year=2020, Use of Degree of Difficulty (DD)=false2021.08 | 74.38 | 73.42 | 0.519 | 0.5103 | 0.8915 | 0.8703 | 0.7472 | — | — | — | — | — | — | — | |
| all-action C3D-LSTMTraining Protocol=All-action, Aggregation Module=LSTM, Feature Extraction Backbone=C3D2018.12 | 61.77 | 67.46 | 49.55 | 36.48 | 84.1 | 73.43 | 0.6478 | — | — | — | — | — | — | — | |
| C3D-LSTMTraining Protocol=Single-action, Aggregation Module=LSTM, Feature Extraction Backbone=C3D2018.12 | 60.47 | 56.36 | 45.93 | 50.29 | 79.12 | 69.27 | 0.6165 | — | — | — | — | — | — | — | |
| C3D-LSTM2020.06 | 60.47 | 56.36 | 45.93 | 50.29 | 79.12 | 69.27 | 0.6165 | — | — | — | — | — | — | — | |
| C3D-LSTM2021.11 | 60.47 | 0.5636 | 0.4593 | 0.5029 | 0.7912 | 0.6927 | 0.6165 | — | — | — | — | — | — | — | |
| C3D-LSTMYear=20172021.08 | 60.47 | 56.36 | 0.4593 | 0.5029 | 0.7912 | 0.6927 | 0.6165 | — | — | — | — | — | — | — | |
| Pose+DCTTraining Protocol=Single-action2018.12 | 53 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Pose+DCT2020.06 | 53 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Pose+DCT2021.11 | 53 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Pose+DCTYear=20142021.08 | 53 | 10 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ST-GCN2020.06 | 32.86 | 57.7 | 16.81 | 12.34 | 66 | 64.83 | 0.4433 | — | — | — | — | — | — | — | |
| ST-GCN2021.11 | 32.86 | 0.577 | 0.1681 | 0.1234 | 0.66 | 0.6483 | 0.4433 | — | — | — | — | — | — | — | |
| ST-GCNYear=20182021.08 | 32.86 | 57.7 | 0.1681 | 0.1234 | 0.66 | 0.6483 | 0.4433 | — | — | — | — | — | — | — |