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


Short Name: LinkBox
Complete name: LinkBox
Description:

LinkBox

Runtime:

10

Submissions:

5

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
51.3828 46.5580 47.1540 29.2760 29.0661 2804 29345 1725 84 1645 76.4136

Detailed
performance:


Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 81.4082 79.3951 69.1663 71.4286 10.2041 169 1972 77 6 83 79.0717
21 28.1157 28.1157 34.5946 0.0000 26.6667 73 599 49 0 49 66.8640
22 23.4058 18.0544 19.1731 8.7379 23.3010 646 6249 540 13 485 68.1127
23 84.7235 79.3628 47.0892 71.4286 14.2857 80 528 77 6 62 81.1275
24 11.7043 1.5702 17.6145 3.3435 61.7021 512 11092 255 34 250 68.8767
25 68.8900 63.6242 77.2736 44.8276 0.5747 297 1511 81 8 71 77.5845
26 51.7473 49.5973 37.9823 35.2941 23.5294 133 1586 76 2 92 76.7176
27 81.5587 75.2595 52.7044 77.4194 6.4516 202 528 207 8 146 77.9335
28 50.2171 42.3213 41.1765 30.0000 20.0000 249 929 83 5 97 71.8336
29 59.3205 57.5163 43.4439 52.0000 16.0000 84 760 66 1 83 75.1668
30 59.6067 59.6067 64.5161 33.3333 4.4444 66 684 51 0 53 77.3521
31 78.9020 78.9020 67.4239 84.4444 4.4444 66 336 63 0 31 78.4280
32 45.2174 44.4617 43.5648 25.0000 35.0000 227 2571 100 1 143 75.0785