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

Short Name: JiaRen.AI
Complete name: JiaRen.AI
Description:

track1_private

Runtime:

na

Submissions:

6

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
56.0474 50.7631 59.6235 40.5037 23.3998 4689 24670 1265 92 1870 77.7520

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 83.4702 81.1216 83.0009 71.4286 4.0816 761 1161 50 7 110 80.6880
21 47.4576 43.4576 65.6322 20.0000 16.6667 123 389 15 1 42 71.6150
22 28.5361 23.1847 32.4510 14.5631 38.8350 217 6559 161 13 309 75.0216
23 90.1873 84.8266 55.2175 85.7143 7.1429 70 327 43 6 33 80.5093
24 15.0994 1.9846 24.7117 5.7751 45.5927 1063 9804 536 44 660 69.1434
25 84.5685 81.2774 91.6036 85.6322 0.0000 140 791 6 5 39 81.1377
26 50.1344 47.9844 48.6267 47.0588 11.7647 609 1168 78 2 149 74.3476
27 81.9130 76.4012 69.7277 87.0968 6.4516 308 517 94 7 100 79.0799
28 45.9929 38.0971 52.4230 55.0000 5.0000 659 628 81 5 97 71.8106
29 54.3138 52.5096 54.0929 52.0000 16.0000 340 636 46 1 83 73.0685
30 61.4221 61.4221 62.3644 33.3333 4.4444 114 578 73 0 84 77.3458
31 81.7151 81.7151 84.1788 86.6667 4.4444 149 240 14 0 35 78.7787
32 60.7561 60.0004 61.7820 25.0000 25.0000 136 1872 68 1 129 78.5207