Weakly-Supervised Action Localization by Generative Attention Modeling

Open paper
Benchmark Model Rank Results
weakly-supervised-action-localization-on-2DGAM#11Mean mAP: 24.4mAP@0.5: 41.0
weakly-supervised-action-localization-on-thumosDGAM#15mAP@0.1:0.7: 37.0mAP@0.1:0.5: 45.6mAP@0.5: 28.8

Rank counts only results with a code link.