 |
Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve) |
| |
Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve) |
Combine ADAM and PRM with PoseFlow
none
1
Benchmark
performance:
| MOTA | MOTP | Total_AP |
|---|---|---|
| 45.3609 | 49.9740 | 63.1687 |
Detailed
performance:
| Video ID | MOTA | MOTP | Total_AP |
|---|---|---|---|
| 20 | 59.9333 | 55.1912 | 79.0873 |
| 21 | 22.9574 | 23.3428 | 31.4631 |
| 22 | -18.4447 | 3.7796 | 30.3391 |
| 23 | 56.6309 | 59.5611 | 73.8443 |
| 24 | 8.8932 | 22.1993 | 39.5985 |
| 25 | 13.6183 | 67.7795 | 57.5629 |
| 26 | 54.7137 | 52.7716 | 57.4284 |
| 27 | 49.6853 | 53.4555 | 74.4306 |
| 28 | 44.8328 | 48.0317 | 51.1759 |
| 29 | 54.5218 | 50.8417 | 62.0612 |
| 30 | 44.8034 | 28.1999 | 54.1826 |
| 31 | 72.3347 | 70.9621 | 82.5843 |
| 32 | 56.0503 | 41.5989 | 59.4529 |