Group activity recognition on Volleyball dataset (test)
93.6MCADIN
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DINBackbone=VGG-162021.08 | 93.6 | 93.8 | — | |
| Lite DINBackbone=VGG-162021.08 | 93.2 | 93.4 | — | |
| DINBackbone=ResNet-182021.08 | 93.1 | 93.3 | — | |
| Lite DINBackbone=ResNet-182021.08 | 92.6 | 92.8 | — | |
| CRMBackbone=ID2021.08 | 92.1 | — | — | |
| PRLBackbone=VGG-162021.08 | 91.4 | 91.8 | — | |
| HiGCINBackbone=ResNet-182021.08 | 91.4 | 92 | — | |
| STBIP*Backbone=Inception-v32021.08 | 91.3 | — | — | |
| ARGBackbone=ResNet-182021.08 | 91.1 | 91.4 | — | |
| SACRFBackbone=ResNet-182021.08 | 90.7 | 91 | — | |
| PCTDMBackbone=ResNet-182021.08 | 90.3 | 90.5 | — | |
| ATBackbone=ResNet-182021.08 | 90 | 90.2 | — | |
| SSUBackbone=Inception-v32021.08 | 89.9 | — | — | |
| SPA+KDBackbone=VGG-162021.08 | 89.3 | 89 | — | |
| stagNetBackbone=VGG-162021.08 | 89.3 | — | — | |
| CERN-2Grouping=2 groups2017.04 | 83.3 | 83.6 | — | |
| CERN-2pooling_style=2 groups2017.04 | 83.3 | 83.6 | — | |
| CERN-2Backbone=VGG-162021.08 | 83.3 | 83.6 | — | |
| B3: CERN-2 w/o p-valuesGrouping=2 groups2017.04 | 82.2 | 82.3 | — | |
| B3: CERN-2 w/o p-valuespooling_style=2 groups, p-values=excluded2017.04 | 82.2 | 82.3 | — | |
| 2-layer LSTMsGrouping=2 groups2017.04 | 81.9 | 82.9 | — | |
| 2-layer LSTMspooling_style=2 groups2017.04 | 81.9 | 82.9 | — | |
| B1: 2-layer LSTMsGrouping=2 groups2017.04 | 80.3 | 80.5 | — | |
| B1: 2-layer LSTMspooling_style=2 groups2017.04 | 80.3 | 80.5 | — | |
| CERN-2Grouping=1 group2017.04 | 73.5 | 72.2 | — | |
| CERN-2pooling_style=1 group2017.04 | 73.5 | 72.2 | — | |
| B3: CERN-2 w/o p-valuesGrouping=1 group2017.04 | 71.7 | 69.8 | — | |
| B3: CERN-2 w/o p-valuespooling_style=1 group, p-values=excluded2017.04 | 71.7 | 69.8 | — | |
| B1: 2-layer LSTMsGrouping=1 group2017.04 | 71.3 | 69.5 | — | |
| B1: 2-layer LSTMspooling_style=1 group2017.04 | 71.3 | 69.5 | — | |
| 2-layer LSTMsGrouping=1 group2017.04 | 70.3 | 65.9 | — | |
| 2-layer LSTMspooling_style=1 group2017.04 | 70.3 | 65.9 | — | |
| SBGARBackbone=Inception-v32021.08 | 66.9 | 67.6 | — | |
| CERN-1Grouping=1 group2017.04 | 34.4 | 34.9 | — | |
| CERN-1pooling_style=1 group2017.04 | 34.4 | 34.9 | — | |
| B2: CERN-1 w/o p-valuesGrouping=1 group2017.04 | 33.3 | 34.3 | — | |
| B2: CERN-1 w/o p-valuespooling_style=1 group, p-values=excluded2017.04 | 33.3 | 34.3 | — | |
| Actor-Transformersbounding boxes=ground-truth2022.07 | — | — | 94.4 | |
| ARGBackbone=Inception-v3, Bounding box source=GT2019.04 | — | — | 92.5 | |
| ARGBackbone=Inception-v3, Bounding box source=PRO2019.04 | — | — | 91.5 | |
| ARGBackbone=VGG16, Bounding box source=GT2019.04 | — | — | 91.9 | |
| ARGBackbone=VGG19, Bounding box source=GT2019.04 | — | — | 92.6 | |
| ARGbounding boxes=predicted2022.07 | — | — | 91.5 | |
| ARGbounding boxes=ground-truth2022.07 | — | — | 92.5 | |
| CERNInput Frames=Multiple Frames2018.12 | — | — | 83.3 | |
| CERNBackbone=VGG162019.04 | — | — | 83.3 | |
| CRMbounding boxes=ground-truth2022.07 | — | — | 93 | |
| DINbounding boxes=ground-truth2022.07 | — | — | 93.6 | |
| Ehsanpour et al.bounding boxes=predicted2022.07 | — | — | 93 | |
| Ehsanpour et al.bounding boxes=ground-truth2022.07 | — | — | 93.1 | |
| GroupFormerbounding boxes=predicted, notes=evaluated with publicly available source codes2022.07 | — | — | 95 | |
| GroupFormerbounding boxes=ground-truth2022.07 | — | — | 95.7 | |
| HDTMInput Frames=Multiple Frames2018.12 | — | — | 81.9 | |
| HDTMBackbone=AlexNet2019.04 | — | — | 81.9 | |
| HGC-Transformerbounding boxes=predicted2022.07 | — | — | 96 | |
| HRNBackbone=VGG192019.04 | — | — | 89.5 | |
| multi-stream convolutional neural network frameworkInput Frames=Multiple Frames2018.12 | — | — | 90.5 | |
| multi-stream convolutional neural network frameworkInput Frames=Single Frame2018.12 | — | — | 86.61 | |
| Pramono et al.bounding boxes=ground-truth2022.07 | — | — | 95 | |
| PRLbounding boxes=ground-truth2022.07 | — | — | 91.4 | |
| Social SceneInput Frames=Multiple Frames2018.12 | — | — | 89.9 | |
| Social SceneInput Frames=Single Frame2018.12 | — | — | 83.8 | |
| SSUBackbone=Inception-v3, Bounding box source=GT2019.04 | — | — | 90.6 | |
| SSUBackbone=Inception-v3, Bounding box source=PRO2019.04 | — | — | 86.2 | |
| SSUbounding boxes=predicted2022.07 | — | — | 86.2 | |
| SSUbounding boxes=ground-truth2022.07 | — | — | 90.6 | |
| stagNetBackbone=VGG16, Bounding box source=GT2019.04 | — | — | 89.3 | |
| stagNetBackbone=VGG16, Bounding box source=PRO2019.04 | — | — | 87.6 | |
| stagNetbounding boxes=predicted2022.07 | — | — | 87.6 | |
| stagNetbounding boxes=ground-truth2022.07 | — | — | 89.3 |