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


Track-4: Person-level Action Recognition in Complex Events

f-mAP@α denotes the frame mAP value calculated under given IOU threshold α; the prefix 'w' indicates the value is calculated with frame weight ; the suffix '@avg' denotes the value is computed by averaging all f-mAP or wf-mAP values under different threshold.

Trackers with "(Baseline)" in the name are submitted by the HiEve team. Results are obtained by running their official code.

Team Name wf-mAP@avg wf-mAP@50 wf-mAP@60 wf-mAP@75 f-mAP@avg f-mAP@50 f-mAP@60 f-mAP@75
MSF
(MM'20 GC Submission)
0.2605 0.3148 0.2895 0.1772 0.3312 0.4044 0.3673 0.2219
L Yuan, Y Zhou, S Chang, Z Huang, Y Chen, YiTu & NUS. Toward Accurate Person-level Action Recognition in Videos of Crowed Scenes. In ACM MM'20
VM
(MM'20 GC Submission)
0.2548 0.3493 0.2922 0.1230 0.2772 0.3616 0.3208 0.1493
Seedland. Seedland. http://www.seedland.cc/
CF
(MM'20 GC Submission)
0.1531 0.1988 0.1797 0.0807 0.2063 0.2645 0.2414 0.1130
Y Hao, Z Liu, H Zhang, City University of Hong Kong. Person-level Action Recognition in Complex Events via TSD-TSM Networks. In ACM MM'20
8A
(MM'20 GC Submission)
0.1509 0.2030 0.1731 0.0766 0.1625 0.2144 0.1877 0.0855
T Li, Z Sun, X Chen, Tencent. Group-Skeleton-Based Human Action Recognition in Complex Events. In ACM MM'20
only_person_rgb
(MM'20 GC Submission)
0.1086 0.1443 0.1254 0.0560 0.1438 0.1801 0.1693 0.0820
Anonymous submission
HWWG
(MM'20 GC Submission)
0.0976 0.1247 0.1094 0.0586 0.1350 0.1662 0.1584 0.0804
Anonymous submission
LittleTry
(MM'20 GC Submission)
0.0970 0.1374 0.1088 0.0448 0.1309 0.1788 0.1507 0.0632
Anonymous submission
RPN+I3D (Baseline) 0.0688 0.0965 0.0791 0.0307 0.0831 0.1101 0.0965 0.0426
R Girdhar, J Carreira, C Doersch, A Zisserman. A Better Baseline for AVA. In CVPR ActivityNet Workshop (AVA Challenge), 2018
qzq-test 0.0660 0.0851 0.0762 0.0368 0.1820 0.2236 0.2004 0.1218
Anonymous submission
FeatureBank (Baseline) 0.0636 0.0869 0.0721 0.0319 0.0842 0.1065 0.0963 0.0497
Chao-Yuan Wu, Christoph Feichtenhofer, Haoqi Fan. Long-Term Feature Banks for Detailed Video Understanding. In CVPR 2019
test_task_4
(MM'20 GC Submission)
0.0598 0.0782 0.0683 0.0328 0.1108 0.1461 0.1262 0.0601
Anonymous submission
SlowFast (Baseline) 0.0529 0.0676 0.0639 0.0271 0.0737 0.0853 0.0900 0.0458
C Feichtenhofer, H Fan, J Malik, K He. SlowFast Networks for Video Recognition. In ICCV, 2019
CVI
(MM'20 GC Submission)
0.0527 0.0758 0.0669 0.0155 0.0644 0.0894 0.0836 0.0202
Anonymous submission
DeepBlueAI
(MM'20 GC Submission)
0.0461 0.0643 0.0524 0.0215 0.0605 0.0774 0.0673 0.0367
Anonymous submission
wantsjean
(MM'20 GC Submission)
0.0420 0.0604 0.0484 0.0171 0.0546 0.0798 0.0625 0.0215
Anonymous submission
newer
(MM'20 GC Submission)
0.0407 0.0633 0.0482 0.0105 0.0518 0.0731 0.0603 0.0221
Anonymous submission
MMIC
(MM'20 GC Submission)
0.0349 0.0509 0.0368 0.0170 0.0382 0.0566 0.0429 0.0152
Anonymous submission
half
(MM'20 GC Submission)
0.0340 0.0501 0.0381 0.0139 0.0523 0.0736 0.0593 0.0240
Anonymous submission
testresultv2
(MM'20 GC Submission)
0.0314 0.0454 0.0360 0.0127 0.0405 0.0520 0.0475 0.0221
Anonymous submission
test_result
(MM'20 GC Submission)
0.0267 0.0449 0.0268 0.0084 0.0260 0.0342 0.0291 0.0147
Anonymous submission
act-2
(MM'20 GC Submission)
0.0243 0.0407 0.0270 0.0052 0.0456 0.0699 0.0533 0.0136
Anonymous submission
Simple Slowfast
(MM'20 GC Submission)
0.0183 0.0243 0.0197 0.0110 0.0197 0.0261 0.0211 0.0118
Anonymous submission
ynnb3
(MM'20 GC Submission)
0.0169 0.0252 0.0174 0.0080 0.0218 0.0280 0.0232 0.0143
YuNan Li, HuiZhou Chen ,XiaoLong Fu. EI:EfficientDet+I3D. 无
txgc
(MM'20 GC Submission)
0.0169 0.0248 0.0179 0.0081 0.0150 0.0201 0.0162 0.0087
Anonymous submission