Challenge and Dataset on Large-scale Human-centric Video Analysis in Complex Events (HiEve)

Short Name: Test
Complete name: Test-pub
Description:

multi fusion

Runtime:

2fps

Submissions:

2

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
46.1385 40.9691 50.7259 23.3998 28.8562 4618 30866 2044 90 2641 75.1760

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 71.0226 68.6740 73.4815 69.3878 6.1224 1410 1958 89 7 236 78.1637
21 18.2453 14.2453 20.4563 0.0000 43.3333 15 750 55 1 57 70.8232
22 16.1430 10.3799 18.8849 4.8544 44.6602 502 7271 367 14 527 69.2445
23 83.9429 78.5822 70.1864 78.5714 7.1429 151 482 87 6 103 76.6640
24 15.1441 2.9235 19.6053 3.3435 55.6231 554 10441 402 41 474 69.0504
25 51.8116 46.5458 61.3326 21.8391 0.5747 425 2258 243 8 248 70.0768
26 49.1129 46.9629 43.2590 29.4118 23.5294 111 1721 61 2 137 76.2772
27 81.4210 79.8462 74.4833 88.7097 0.0000 372 470 102 2 177 79.0055
28 30.8330 22.9372 36.4502 15.0000 35.0000 380 1243 129 5 136 69.4603
29 49.7988 47.9946 47.6294 28.0000 24.0000 144 927 52 1 134 76.0231
30 48.3611 48.3611 47.4260 15.5556 0.0000 34 892 98 0 95 77.0562
31 79.9002 79.9002 78.5926 84.4444 2.2222 193 224 26 0 65 75.5905
32 45.3875 43.1204 42.1526 22.5000 25.0000 327 2229 333 3 252 75.5341