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

Short Name: Spatio temporal poin
Complete name: STPP
Description:

STPP

Runtime:

2000

Submissions:

1

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
37.4682 32.0690 40.1731 20.3568 29.8006 7395 31638 4536 94 3486 75.6045

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 69.8575 67.5089 59.3955 63.2653 6.1224 1188 2053 355 7 370 78.9372
21 13.1605 9.1605 25.5913 0.0000 26.6667 202 559 110 1 69 66.6063
22 6.9228 0.7481 13.4758 4.8544 26.2136 1182 6556 1297 15 895 68.3234
23 77.2302 70.9760 57.6674 64.2857 14.2857 93 727 201 7 172 80.2998
24 4.7502 -7.1723 12.0422 1.5198 62.6140 820 11273 700 40 415 68.5588
25 54.4796 49.2138 67.7478 26.4368 0.0000 661 1891 212 8 195 73.4527
26 30.1613 28.0113 31.8634 35.2941 23.5294 717 1717 164 2 161 76.3047
27 67.1718 61.6600 38.4851 58.0645 3.2258 205 914 549 7 398 78.2544
28 5.0533 -2.8425 27.3478 20.0000 25.0000 1148 1007 250 5 167 71.2488
29 17.7917 15.9875 32.6290 20.0000 32.0000 644 1048 147 1 147 65.7519
30 48.7645 48.7645 51.7782 22.2222 2.2222 88 736 192 0 132 76.8944
31 67.4229 67.4229 57.0438 62.2222 4.4444 116 485 117 0 131 78.8978
32 38.6578 37.9021 44.0587 22.5000 40.0000 331 2672 242 1 234 75.3102