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

Short Name: SiamMOT
Complete name: SiamMOT
Description:

n / a

Runtime:

n /a

Submissions:

6

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
53.1683 48.4584 51.6965 26.6527 27.4921 2837 28485 1308 82 1730 79.2036

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 81.0981 79.0850 72.0709 87.7551 4.0816 995 1170 90 6 123 82.0301
21 29.6112 25.6112 40.5694 3.3333 36.6667 38 639 29 1 34 69.9096
22 26.2079 20.4448 26.9651 11.6505 36.8932 206 6707 250 14 452 74.3778
23 81.2444 77.6706 54.6394 71.4286 14.2857 55 703 83 4 77 81.5809
24 17.4075 7.2734 24.5229 6.0790 54.7112 259 10614 220 34 247 73.8841
25 56.5382 51.2724 69.2025 12.6437 3.4483 117 2481 41 8 141 76.7810
26 57.1237 54.9737 49.3940 47.0588 17.6471 227 1314 54 2 90 79.7856
27 74.3161 68.8043 46.8414 77.4194 4.8387 230 815 260 7 233 79.1416
28 57.4812 51.1646 43.2234 45.0000 15.0000 284 709 84 4 95 74.5261
29 54.8502 53.0460 54.2857 44.0000 12.0000 347 621 42 1 64 77.9603
30 62.7332 62.7332 66.1650 37.7778 2.2222 24 668 47 0 53 81.5974
31 85.7532 85.7532 71.3981 88.8889 4.4444 34 229 51 0 44 80.5956
32 64.2155 63.4598 58.9574 32.5000 20.0000 21 1815 57 1 77 80.8296