Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve) |
LSTC: Boosting Atomic Action Detection with Long-S
none
1
Benchmark
performance:
F_w_mAP | F_w_mAP_50 | F_w_mAP_60 | F_w_mAP_75 | F_mAP | F_mAP_50 | F_mAP_60 | F_mAP_75 |
---|---|---|---|---|---|---|---|
0.0744 | 0.0967 | 0.0853 | 0.0412 | 0.0890 | 0.1136 | 0.1054 | 0.0481 |
Detailed
performance:
Video ID | F_w_mAP | F_w_mAP_50 | F_w_mAP_60 | F_w_mAP_75 | F_mAP | F_mAP_50 | F_mAP_60 | F_mAP_75 |
---|---|---|---|---|---|---|---|---|
20 | 0.0325 | 0.0395 | 0.0361 | 0.0219 | 0.0494 | 0.0606 | 0.0531 | 0.0345 |
21 | 0.0312 | 0.0555 | 0.0292 | 0.0088 | 0.0281 | 0.0514 | 0.0265 | 0.0065 |
22 | 0.0927 | 0.1176 | 0.1079 | 0.0527 | 0.0997 | 0.1332 | 0.1134 | 0.0527 |
23 | 0.2702 | 0.3194 | 0.3046 | 0.1867 | 0.3229 | 0.3768 | 0.3659 | 0.2260 |
24 | 0.0422 | 0.0647 | 0.0467 | 0.0151 | 0.0469 | 0.0685 | 0.0521 | 0.0201 |
25 | 0.0049 | 0.0077 | 0.0053 | 0.0016 | 0.0060 | 0.0090 | 0.0065 | 0.0025 |
26 | 0.1804 | 0.1985 | 0.1911 | 0.1515 | 0.2459 | 0.2598 | 0.2609 | 0.2172 |
27 | 0.2373 | 0.2903 | 0.2695 | 0.1520 | 0.2388 | 0.2930 | 0.2700 | 0.1535 |
28 | 0.0298 | 0.0549 | 0.0232 | 0.0112 | 0.0282 | 0.0562 | 0.0163 | 0.0121 |
29 | 0.1569 | 0.2015 | 0.1848 | 0.0845 | 0.1760 | 0.2313 | 0.2056 | 0.0910 |
30 | 0.1237 | 0.1422 | 0.1468 | 0.0822 | 0.1187 | 0.1260 | 0.1360 | 0.0942 |
31 | 0.3459 | 0.5941 | 0.3201 | 0.1234 | 0.3624 | 0.6149 | 0.3430 | 0.1292 |
32 | 0.0484 | 0.0841 | 0.0449 | 0.0161 | 0.0527 | 0.0944 | 0.0471 | 0.0165 |