HISAN (VGG-16)

Reported on 6 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Time Series6 results

Action DetectiononUCF101-24
Frame-mAP 0.5
73.71
best: 90.3 (STAR/L)
Action DetectiononUCF101-24
Video-mAP 0.2
80.42
best: 88.8 (HIT)
Action DetectiononUCF101-24
Video-mAP 0.5
49.5
best: 76.3 (Stable Mean Teacher (I3D))
Action DetectiononJ-HMDB
Frame-mAP 0.5
76.72
best: 88.5 (SiA)
Action DetectiononJ-HMDB
Video-mAP 0.2
85.97
best: 89.7 (HIT)
Action DetectiononJ-HMDB
Video-mAP 0.5
84.02
best: 88.1 (HIT)