RGNet: A Unified Clip Retrieval and Grounding Network for Long Videos

Benchmark Model Rank Results
natural-language-moment-retrieval-on-madRGNet#3R@1,IoU=0.1: 12.43R@1,IoU=0.3: 9.48R@1,IoU=0.5: 5.61
natural-language-queries-on-ego4dRGNet#4R@1 Mean(0.3 and 0.5): 16.55R@1 IoU=0.3: 20.63R@1 IoU=0.5: 12.47