| skeleton-based-action-recognition-on-kinetics-skeleton | AGCN (2-stream) | #20 | Accuracy: 36.9GFLOPS per prediction: 26.91 |
| skeleton-based-action-recognition-on-kinetics-skeleton | AGCN (1-stream) | #22 | Accuracy: 35GFLOPS per prediction: 13.45 |
| skeleton-based-action-recognition-on-kinetics-skeleton | S-TR (2-stream) | #25 | Accuracy: 34.7GFLOPS per prediction: 23.24 |
| skeleton-based-action-recognition-on-kinetics-skeleton | ST-GCN (2-stream) | #26 | Accuracy: 34.4GFLOPS per prediction: 24.09 |
| skeleton-based-action-recognition-on-kinetics-skeleton | ST-GCN (1-stream) | #27 | Accuracy: 33.4GFLOPS per prediction: 12.04 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoST-GCN (2-stream) | #28 | Accuracy: 33.1GFLOPS per prediction: 0.32 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoAGCN (1-stream) | #29 | Accuracy: 33GFLOPS per prediction: 0.18 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoS-TR (2-stream) | #30 | Accuracy: 32.7GFLOPS per prediction: 0.31 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoST-GCN* (2-stream) | #31 | Accuracy: 32.2GFLOPS per prediction: 0.22 |
| skeleton-based-action-recognition-on-kinetics-skeleton | S-TR (1-stream) | #32 | Accuracy: 32GFLOPS per prediction: 11.62 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoST-GCN (1-stream) | #33 | Accuracy: 31.8GFLOPS per prediction: 0.16 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoST-GCN* (1-stream) | #36 | Accuracy: 30.2GFLOPS per prediction: 0.11 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoS-TR* (2-stream) | #37 | Accuracy: 29.9GFLOPS per prediction: 0.22 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoS-TR (1-stream) | #38 | Accuracy: 29.7 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoAGCN* (2-stream) | #39 | Accuracy: 27.5GFLOPS per prediction: 0.25 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoS-TR* (1-stream) | #40 | Accuracy: 27.4GFLOPS per prediction: 0.11 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoAGCN* (1-stream) | #41 | Accuracy: 23.3GFLOPS per prediction: 0.12 |
| skeleton-based-action-recognition-on-kinetics-skeleton | CoAGCN (2-stream) | #42 | GFLOPS per prediction: 0.36 |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | S-TR (2-stream) | #44 | Accuracy (Cross-Subject): 84.8Accuracy (Cross-Setup): 86.2… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoS-TR* (2-stream) | #45 | Accuracy (Cross-Subject): 84.8Accuracy (Cross-Setup): 86.1… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoST-GCN* (2-stream) | #48 | Accuracy (Cross-Subject): 84.0Accuracy (Cross-Setup): 85.5… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | AGCN (2-stream) | #49 | Accuracy (Cross-Subject): 84Accuracy (Cross-Setup): 85.4… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | ST-GCN (2-stream) | #50 | Accuracy (Cross-Subject): 83.7Accuracy (Cross-Setup): 85.1… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoAGCN* (2-stream) | #55 | Accuracy (Cross-Subject): 80.4Accuracy (Cross-Setup): 82… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | S-TR (1-stream) | #56 | Accuracy (Cross-Subject): 80.2Accuracy (Cross-Setup): 81.8… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoS-TR* (1-stream) | #57 | Accuracy (Cross-Subject): 79.7Accuracy (Cross-Setup): 81.7… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | AGCN (1-stream) | #58 | Accuracy (Cross-Subject): 79.7Accuracy (Cross-Setup): 80.7… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoST-GCN* (1-stream) | #59 | Accuracy (Cross-Subject): 79.4Accuracy (Cross-Setup): 81.6… |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | ST-GCN (1-stream) | #60 | Accuracy (Cross-Subject): 79GFLOPS per prediction: 16.73 |
| skeleton-based-action-recognition-on-ntu-rgbd-120 | CoAGCN* (1-stream) | #61 | Accuracy (Cross-Subject): 77.3Accuracy (Cross-Setup): 79.1… |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoS-TR* (2-stream) | #65 | Accuracy (CS): 88.9Accuracy (CV): 94.8GFLOPs per pred: 0.3 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoST-GCN* (2-stream) | #67 | Accuracy (CS): 88.3Accuracy (CV): 95GFLOPs per pred: 0.32 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoST-GCN* | #77 | Accuracy (CS): 86.3Accuracy (CV): 93.8GFLOPs per pred: 0.16 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoS-TR* | #78 | Accuracy (CS): 86.3Accuracy (CV): 92.4GFLOPs per pred: 0.15 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | ST-GCN | #79 | Accuracy (CS): 86Accuracy (CV): 93.4GFLOPs per pred: 16.73 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoAGCN* (2-stream) | #80 | Accuracy (CS): 86.0Accuracy (CV): 93.1GFLOPs per pred: 0.44 |
| skeleton-based-action-recognition-on-ntu-rgbd-60 | CoAGCN* | #84 | Accuracy (CS): 84.1Accuracy (CV): 92.6 |