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

Short Name: FairerMOT
Complete name: a fairer way to track
Description:

……

Runtime:

1000

Submissions:

3

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
36.1134 31.0588 39.8582 17.1039 50.5771 6291 36439 1783 88 2311 74.6581

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 67.3680 65.0194 63.5518 61.2245 4.0816 1539 2177 177 7 293 77.6261
21 6.3809 2.3809 11.6918 0.0000 86.6667 25 902 12 1 35 68.8384
22 15.3910 10.8629 16.0527 0.9709 46.6019 360 7477 376 11 490 69.1816
23 85.7047 80.3440 58.7101 71.4286 14.2857 101 464 76 6 80 76.3086
24 12.0095 -0.5091 13.6157 2.7356 57.1429 625 10771 422 42 439 66.9192
25 -27.6186 -30.9097 2.5765 0.0000 99.4253 1815 5886 48 5 37 59.9379
26 47.4462 45.2962 42.4929 35.2941 23.5294 222 1661 72 2 121 77.0293
27 77.1305 72.4061 58.3923 75.8065 4.8387 420 543 199 6 181 78.5517
28 28.7801 19.3052 31.5173 10.0000 45.0000 288 1426 90 6 135 67.0811
29 50.5588 48.7546 51.2115 40.0000 24.0000 101 943 62 1 116 76.4951
30 39.1831 39.1831 43.6031 11.1111 6.6667 121 1023 62 0 100 74.5322
31 70.6443 70.6443 71.5278 80.0000 4.4444 255 343 49 0 65 73.3600
32 36.1059 35.3502 39.0166 17.5000 40.0000 419 2823 138 1 219 69.2011