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


Short Name: LCL
Complete name: LCL
Description:

LCL

Runtime:

LCL

Submissions:

3

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
14.5562 10.4206 23.5454 5.6663 68.7303 6920 49869 2744 72 1980 41.9371

Detailed
performance:


Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 69.6060 67.2574 58.0241 63.2653 4.0816 1205 2078 343 7 389 78.8513
21 11.4656 7.4656 25.5114 0.0000 26.6667 203 547 138 1 73 66.5325
22 7.3246 1.1499 12.8251 6.7961 23.3010 1167 6546 1283 15 883 68.5380
23 76.0482 69.7940 58.0823 71.4286 14.2857 85 720 269 7 179 80.2461
24 5.0182 -6.6062 11.9029 1.8237 62.9179 813 11242 702 39 425 68.6654
25 -10.4908 -11.1490 3.3067 0.0000 91.3793 907 5793 9 1 18 62.8915
26 -17.5806 -17.5806 0.0000 0.0000 100.0000 654 3720 0 0 0 0.0000
27 -8.0890 -8.0890 0.0364 0.0000 100.0000 411 5081 0 0 0 0.0000
28 -10.0276 -13.1859 2.7901 0.0000 95.0000 330 2457 0 2 12 56.2913
29 -13.3214 -13.3214 0.0000 0.0000 100.0000 298 2237 0 0 0 0.0000
30 -15.7337 -15.7337 0.6800 0.0000 100.0000 341 1954 0 0 1 63.1662
31 -7.3956 -7.3956 0.0000 0.0000 100.0000 163 2204 0 0 0 0.0000
32 -6.4839 -6.4839 0.0000 0.0000 100.0000 343 5290 0 0 0 0.0000