Finding Action Tubes

Open paper
Benchmark Model Rank Results
skeleton-based-action-recognition-on-j-hmdbAction Tubes#4Accuracy (RGB+pose): 62.5
action-detection-on-j-hmdbAction Tubes#8Frame-mAP 0.5: 36.2Video-mAP 0.5: 53.3

Rank counts only results with a code link.