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

Short Name: SMOT1
Complete name: SMOT1
Description:

n/a

Runtime:

n/a

Submissions:

3

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
48.1823 43.4149 42.2015 19.0976 34.1028 995 33363 1746 83 1638 78.7917

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 74.7946 72.7815 60.9610 59.1837 12.2449 57 2815 135 6 120 81.6788
21 30.3091 26.3091 34.8518 3.3333 46.6667 24 647 28 1 27 69.8734
22 23.6221 18.2707 19.0199 9.7087 42.7184 171 6872 371 13 374 75.0705
23 75.6467 70.2860 42.1508 50.0000 14.2857 55 933 104 6 100 78.1170
24 15.2409 5.7029 20.2690 3.6474 62.3100 80 11105 199 32 180 74.5334
25 48.5178 43.2520 57.3504 6.3218 9.1954 100 2884 142 8 128 76.4242
26 44.2473 42.0973 29.9039 23.5294 35.2941 78 1900 96 2 99 79.1376
27 70.2814 64.7696 36.9357 59.6774 8.0645 148 1029 333 7 268 78.6385
28 44.5717 36.6759 32.7078 20.0000 30.0000 182 1145 77 5 94 73.4700
29 54.6714 52.8672 43.5355 28.0000 24.0000 59 890 65 1 69 78.2293
30 56.8835 56.8835 60.1948 31.1111 2.2222 8 791 56 0 47 81.9875
31 80.9437 80.9437 60.3856 82.2222 4.4444 14 325 81 0 64 80.9879
32 60.2079 58.6965 57.9795 22.5000 30.0000 19 2027 59 2 68 81.0768