ResearchBenchmarksDense Captioning on VG 1.2 (test)Follow19.7APDetCLIPv39.61212.23114.8517.469Apr 14, 2024Evaluation ResultsMethodMethodLinksAPDetCLIPv3Backbone=Swin-LBackbone=Swin-L2024.0419.7DetCLIPv3Backbone=Swin-TBackbone=Swin-T2024.0418.4GRIT2024.0415.5CapDet2024.0415.4TDC+ROCSU2024.0411.9CAG-Net2024.0410.5COCG2024.0410.4JIVC2024.0410