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

Short Name: Adaptive FairMOT
Complete name: Adaptive FairMOT
Description:

Adaptive FairMOT

Runtime:

5

Submissions:

11

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
60.2282 54.6567 60.0026 41.2382 20.2518 4530 21292 1889 97 1808 74.9261

Detailed
performance:

Video ID MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
20 85.9095 83.5609 86.9547 79.5918 4.0816 720 924 37 7 50 81.5332
21 43.8684 39.8684 53.3654 10.0000 20.0000 91 433 39 1 29 67.2793
22 30.2462 23.6598 25.7250 15.5340 13.5922 1300 4714 757 16 664 67.6156
23 87.9572 82.5965 70.2147 78.5714 7.1429 196 314 30 6 53 72.5310
24 19.5443 6.4295 19.0592 7.2948 46.5046 433 10012 361 44 418 70.5633
25 77.1245 72.5170 75.4490 74.1379 0.0000 253 927 209 7 56 73.8690
26 69.6505 67.5005 51.3929 64.7059 0.0000 415 640 74 2 91 69.9405
27 87.5418 82.0300 74.9632 87.0968 3.2258 290 263 80 7 51 77.5273
28 63.7979 60.6396 65.6941 45.0000 20.0000 228 633 56 2 72 75.8698
29 71.3456 69.5414 71.3874 72.0000 16.0000 151 459 31 1 48 78.1330
30 66.5658 66.5658 68.8908 44.4444 0.0000 68 527 68 0 85 75.5698
31 91.8784 91.8784 90.4420 95.5556 2.2222 38 125 16 0 22 75.6175
32 65.9924 62.9695 63.6477 40.0000 15.0000 347 1321 131 4 169 72.7685