QVHighlights: Detecting Moments and Highlights in Videos via Natural Language Queries

Benchmark Model Rank Results
highlight-detection-on-qvhighlightsMoment-DETR w/ PT#18mAP: 37.43Hit@1: 60.17
moment-retrieval-on-charades-staMoment-DETR w/ PT (on 10K HowTo100M videos)#20R@1 IoU=0.5: 55.65R@1 IoU=0.7: 34.17
moment-retrieval-on-charades-staMoment-DETR#21R@1 IoU=0.5: 53.63R@1 IoU=0.7: 31.37
moment-retrieval-on-qvhighlightsMoment-DETR (w/ PT ASR Cpations)#26mAP: 36.14R@1 IoU=0.5: 59.78R@1 IoU=0.7: 40.33mAP@0.5: 60.51