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


Short Name: DeepBlueAI
Complete name: DeepBlueAI
Description: refine
Runtime: 20min Hz
Submissions: 4

Benchmark
performance:

T_w_AP T_w_AP_50 T_w_AP_75 T_w_AP_90 T_AP T_AP_50 T_AP_75 T_AP_90
52.0535 62.7362 49.2952 44.1291 56.3326 66.8391 53.8037 48.3551

Detailed
performance:


Video Name T_w_AP T_w_AP_50 T_w_AP_75 T_w_AP_90 T_AP T_AP_50 T_AP_75 T_AP_90
hm_in_waiting_hall (ID:20) 58.9968 73.5504 53.8600 49.5801 59.1343 73.6028 54.0751 49.7250
hm_in_bus (ID:21) 45.6038 41.7160 48.8151 46.2802 47.2667 44.2199 50.3226 47.2577
hm_in_dining_room2 (ID:22) 21.8818 28.8677 19.1101 17.6677 22.0636 29.0083 19.4019 17.7807
hm_in_lab2 (ID:23) 68.1854 80.3144 65.8914 58.3504 70.1655 80.8580 68.7232 60.9154
hm_in_subway_station (ID:24) 41.1491 47.7876 39.2678 36.3919 41.8398 48.6109 39.8052 37.1034
hm_in_passage (ID:25) 80.9858 79.2183 83.5384 80.2007 82.0901 79.4371 84.3794 82.4538
hm_in_fighting4 (ID:26) 58.9653 72.4573 56.3633 48.0752 62.3549 74.7570 60.4720 51.8357
hm_in_shopping_mall3 (ID:27) 60.9672 76.2562 58.7194 47.9259 62.6631 78.1210 60.3848 49.4836
hm_in_restaurant (ID:28) 62.4754 66.6928 63.2125 57.5208 62.8245 65.6454 63.8166 59.0115
hm_in_accident (ID:29) 56.1331 72.1734 50.4348 45.7912 57.4997 72.6825 52.2838 47.5329
hm_in_stair3 (ID:30) 44.4359 55.9097 40.5950 36.8031 42.9929 54.6293 38.4969 35.8526
hm_in_crossroad (ID:31) 74.0376 86.5329 72.8813 62.6985 74.4435 86.0488 71.9640 65.3176
hm_in_robbery (ID:32) 46.5877 54.1683 44.9701 40.6247 47.2122 54.4431 45.8502 41.3433