|
Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve) |
FasterRCNN(Baseline)
none
1
Benchmark
performance:
| MR | AP | JI |
|---|---|---|
| 0.6722 | 0.5935 | 0.5156 |
Detailed
performance:
| Video ID | MR | AP | JI |
|---|---|---|---|
| 20 | 0.5690 | 0.8882 | 0.7069 |
| 21 | 0.9066 | 0.4413 | 0.3516 |
| 22 | 0.9669 | 0.4317 | 0.3402 |
| 23 | 0.1418 | 0.8935 | 0.8570 |
| 24 | 0.9468 | 0.2070 | 0.2237 |
| 25 | 0.9988 | 0.0069 | 0.0320 |
| 26 | 0.5657 | 0.6087 | 0.5003 |
| 27 | 0.3651 | 0.8715 | 0.8013 |
| 28 | 0.6894 | 0.6707 | 0.5360 |
| 29 | 0.7079 | 0.6482 | 0.5471 |
| 30 | 0.6422 | 0.6785 | 0.6019 |
| 31 | 0.4464 | 0.8838 | 0.7797 |
| 32 | 0.7922 | 0.4854 | 0.4254 |