Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ShanghaiTech

Benchmarks

Task NameDataset NameSOTA ResultTrend
Crowd CountingShanghaiTech Part A (test)
MAE46
271
Video Anomaly DetectionShanghaiTech (test)
AUC0.9814
211
Crowd CountingShanghaiTech Part B (test)
MAE5.4
208
Crowd CountingShanghaiTech Part B
MAE6.2
177
Crowd CountingShanghaiTech Part A
MAE52.74
155
Video Anomaly DetectionShanghaiTech (SHT) (test)
Frame-level AUC98.1
82
Anomaly DetectionShanghaiTech
AUROC0.902
68
Video Anomaly DetectionShanghaiTech
Micro AUC0.867
51
Video Anomaly DetectionShanghaiTech standard (test)
Frame-Level AUC97.98
50
Anomaly DetectionShanghaiTech (test)
RBDC47.73
34
Line Segment DetectionShanghaiTech (test)
F^H87.2
32
Video Anomaly DetectionShanghaiTech (SHTech) (test)
AUROC0.859
24
Video Anomaly DetectionShanghaiTech (SHT)
AUC80.5
20
Anomaly DetectionShanghaiTech
Micro AUC (Frame)83.6
20
Anomaly LocalizationShanghaiTech
Pixel AUROC92.6
18
Video Anomaly DetectionShanghaiTech classic (test)
AUC76.6
16
Line Segment DetectionShanghaiTech v1.0 (test)
sAP564.3
15
Video Anomaly DetectionShanghaiTech Mixture [01,02] scene-dependent
Micro AUC91
14
Crowd CountingShanghaiTech-A -> UCF-QNRF (test)
MAE105.3
13
Video Anomaly DetectionShanghaiTech Campus
RBDC62.1
12
Frame-level Video Anomaly DetectionShanghaiTech
AUC0.8528
11
Skeleton-based Video Anomaly DetectionShanghaiTech HR (test)
AUC88.1
10
Crowd CountingShanghaiTech B 12 (test)
MAE5.8
10
Video Anomaly DetectionShanghaiTech reorganization (test)
AUC0.9761
9
Pedestrian Anomaly DetectionShanghaiTech
AUC0.76
8
Showing 25 of 41 rows