ResearchBenchmarksVisual Grounding on Flickr30k EntitiesFollow77.5AccuracyGiT-H25.81239.23152.6566.069May 9, 2018Apr 30, 2019Apr 20, 2020Apr 11, 2021Apr 2, 2022Mar 24, 2023Mar 14, 2024Evaluation ResultsMethodMethodLinksAccuracyGiT-HBackbone=Huge, Trainin...Backbone=Huge, Training=Universal2024.0377.5DDPNBackbone=ResNet-101Backbone=ResNet-1012018.0573.3GiT-LBackbone=Large, Traini...Backbone=Large, Training=Universal2024.0372.7GiT-BBackbone=Base, Trainin...Backbone=Base, Training=Universal2024.0371DDPNBackbone=VGG-16Backbone=VGG-162018.0570QRC2018.0565.1MCB2018.0548.7GroundeR2018.0547.8SCRC2018.0527.8