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

Short Name: FairMOT(baseline)
Complete name: FairMOT(baseline)
Description:

FairMOT(baseline)

Runtime:

8000

Submissions:

2

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
35.0298 30.4922 46.6524 16.2644 44.1763 6523 37750 995 79 2312 75.5656

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 31.7687 29.7556 56.3191 46.9388 16.3265 4315 3757 68 6 271 76.9033
21 6.8794 2.8794 12.7971 0.0000 90.0000 7 919 8 1 17 66.2088
22 15.8236 11.7071 20.4334 0.9709 64.0777 101 7887 183 10 345 73.4748
23 84.5897 79.2290 68.0951 71.4286 14.2857 87 561 43 6 70 76.1335
24 8.7931 -1.0430 15.2357 1.2158 70.5167 296 11760 194 33 321 72.9207
25 24.3083 19.0425 41.4994 0.5747 25.2874 461 3988 147 8 335 73.9949
26 47.3925 45.2425 46.6415 29.4118 29.4118 152 1756 49 2 155 74.8436
27 81.8933 77.1689 77.2451 82.2581 3.2258 289 563 68 6 196 80.0586
28 32.4911 24.5953 37.9028 5.0000 45.0000 131 1535 44 5 139 70.4512
29 54.9397 53.1355 53.1897 32.0000 24.0000 59 912 37 1 97 76.6206
30 36.8633 36.8633 48.7885 17.7778 6.6667 141 1053 58 0 107 77.0658
31 76.3612 76.3612 84.1360 80.0000 4.4444 164 336 21 0 54 74.8192
32 41.0586 40.3029 45.7380 17.5000 37.5000 320 2723 75 1 205 69.3118