TACNet: Transition-Aware Context Network for Spatio-Temporal Action Detection
Open paper
Benchmark
Model
Rank
Results
action-detection-on-j-hmdb
TACNet
–
Frame-mAP 0.5: 65.5
Video-mAP 0.2: 74.1
Video-mAP 0.5: 73.4
action-detection-on-ucf101-24
TACNet
–
Frame-mAP 0.5: 72.1
Video-mAP 0.2: 77.5
Video-mAP 0.5: 52.9
Rank counts only results with a code link.