| action-classification-on-toyota-smarthome | π-ViT | #1 | CS: 72.9CV1: 55.2CV2: 64.8 |
| action-recognition-in-videos-on-ntu-rgbd-120 | π-ViT (RGB + Pose) | #3 | Accuracy (Cross-Setup): 96.1Accuracy (Cross-Subject): 95.1 |
| action-recognition-in-videos-on-ntu-rgbd-120 | π-ViT (RGB only) | #7 | Accuracy (Cross-Setup): 91.9Accuracy (Cross-Subject): 92.9 |
| action-recognition-in-videos-on-ntu-rgbd-60 | π-ViT (RGB + Pose) | #3 | Accuracy (CS): 96.3Accuracy (CV): 99.0 |
| action-recognition-in-videos-on-ntu-rgbd-60 | π-ViT (RGB only) | #9 | Accuracy (CS): 94.0Accuracy (CV): 97.9 |