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


Short Name: DeepBlueAI
Complete name: DeepBlueAI
Description: Last
Runtime: 20min Hz
Submissions: 1

Benchmark
performance:

MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
38.5892 33.2474 47.6724 32.5289 24.6590 13040 27194 2554 93 2767 75.8830

Detailed
performance:


Video Name MOTA w_MOTA IDF1 MT ML FP FN ID_Sw ID_Sw_DT Frag MOTP
hm_in_waiting_hall (ID:20) 68.5666 66.5535 68.8336 73.4694 6.1224 2127 1528 95 6 181 78.6986
hm_in_bus (ID:21) 5.1844 1.1844 32.4182 10.0000 20.0000 376 488 87 1 68 66.8237
hm_in_dining_room2 (ID:22) 5.9854 -0.6010 20.9451 13.5922 13.5922 2517 5679 930 16 842 67.5613
hm_in_lab2 (ID:23) 82.9839 78.5166 52.9576 71.4286 14.2857 174 520 69 5 94 81.1302
hm_in_subway_station (ID:24) 6.9987 -4.9238 19.1634 3.9514 55.3191 1294 10800 397 40 416 68.4256
hm_in_passage (ID:25) 51.5316 46.2658 70.6253 55.7471 0.0000 1505 1254 184 8 155 77.1187
hm_in_fighting4 (ID:26) 22.1505 20.0005 32.4623 35.2941 23.5294 1306 1513 77 2 164 76.0458
hm_in_shopping_mall3 (ID:27) 75.0640 69.5522 64.3494 85.4839 3.2258 551 500 216 7 219 77.5973
hm_in_restaurant (ID:28) 0.0395 -9.4354 32.8554 30.0000 15.0000 1564 841 127 6 143 71.8010
hm_in_accident (ID:29) 30.0402 28.2360 38.6743 48.0000 12.0000 777 695 93 1 116 74.5005
hm_in_stair3 (ID:30) 56.4801 56.4801 64.6110 26.6667 2.2222 164 595 104 0 108 76.4925
hm_in_crossroad (ID:31) 73.2759 73.2759 74.1583 84.4444 4.4444 218 319 52 0 58 78.1741
hm_in_robbery (ID:32) 42.3062 41.5505 48.0140 25.0000 32.5000 467 2462 123 1 203 74.9733