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

Short Name: lalalala
Complete name: lalalala
Description:

none

Runtime:

none

Submissions:

2

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
0.6865 0.1882 0.6836 1.1878 0.6087 0.1407 0.5941 1.0913

Detailed
performance:

Video ID 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
20 0.5933 0.1518 0.6317 0.9963 0.5957 0.1496 0.6351 1.0023
21 1.9838 0.3750 2.0642 3.5121 1.8904 0.3044 1.9334 3.4334
22 0.7901 0.4256 0.7246 1.2200 0.7974 0.4222 0.7528 1.2173
23 2.8421 0.5797 2.7782 5.1685 2.8394 0.5712 2.7728 5.1741
24 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
25 1.2402 0.5064 1.5010 1.7131 1.2289 0.5116 1.4014 1.7738
26 1.2560 0.0023 0.8416 2.9240 1.5822 0.0013 1.2017 3.5436
27 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
28 4.2302 1.6929 4.1074 6.8903 3.4564 1.3820 3.1863 5.8010
29 0.5950 0.0163 0.4248 1.3440 0.6286 0.0116 0.4117 1.4626
30 10.8466 6.5332 11.9113 14.0952 11.6680 6.5450 12.6869 15.7720
31 0.1335 0.0304 0.1052 0.2648 0.1721 0.0320 0.1324 0.3518
32 0.0330 0.0000 0.0153 0.0837 0.0290 0.0000 0.0129 0.0741