List your company

Image-Text Retrieval on MS-COCO 1K (test)

84.9Image-to-Text R@1CDDS
74.60477.27779.9582.623Mar 5, 2026

Evaluation results

MethodLinks
CDDSBackbone=Swin, Resolution=384x384, Number of patches=12x122026.0384.99899.673.594.698548.6
CDDSBackbone=Swin, Resolution=224x224, Number of patches=7x72026.0384.697.999.472.193.597.6545.1
LAPSBackbone=Swin, Resolution=384x384, Number of patches=12x122026.0384.197.499.272.193.997.4544.1
LAPSBackbone=Swin, Resolution=224x224, Number of patches=7x72026.038497.699.372.193.797.3544.1
VSE++Backbone=Swin, Resolution=224x224, Number of patches=7x72026.0383.397.599.3719396.7540.9
CHANBackbone=Swin, Resolution=384x384, Number of patches=12x122026.0383.197.399.270.493.197.1540.2
VSE++Backbone=Swin, Resolution=384x384, Number of patches=12x122026.0382.997.799.471.393.597.3542.1
SGRBackbone=Swin, Resolution=384x384, Number of patches=12x122026.0381.996.799.169.392.896.7536.6
CHANBackbone=Swin, Resolution=224x224, Number of patches=7x72026.0381.697.299.370.693.797.6539.8
SCANBackbone=Swin, Resolution=384x384, Number of patches=12x122026.0381.696.899.169.192.796.7536.1
SGRBackbone=Swin, Resolution=224x224, Number of patches=7x72026.0381.297.199.169.993.297.2537.7
SCANBackbone=Swin, Resolution=224x224, Number of patches=7x72026.0380.99799.169.793.197.1536.9
CDDSBackbone=ViT, Resolution=384x384, Number of patches=24x242026.0379.196.898.968.392.696.9532.6
CDDSBackbone=ViT, Resolution=224x224, Number of patches=14x142026.037995.998.766.591.896.8528.7
LAPSBackbone=ViT, Resolution=224x224, Number of patches=14x142026.0378.795.598.366.291.396.2526.3
LAPSBackbone=ViT, Resolution=384x384, Number of patches=24x242026.0378.796.398.96892.496.8531
CHANBackbone=ViT, Resolution=384x384, Number of patches=24x242026.0378.195.898.666.192.196.6527.3
SGRBackbone=ViT, Resolution=224x224, Number of patches=14x142026.0377.2959865.190.795.8521.8
CHANBackbone=ViT, Resolution=224x224, Number of patches=14x142026.0377.195.198.1659196522.2
VSE++Backbone=ViT, Resolution=384x384, Number of patches=24x242026.037795.798.464.691.196.2523
SCANBackbone=ViT, Resolution=384x384, Number of patches=24x242026.0376.195.598.565.191.696.3523.1
SCANBackbone=ViT, Resolution=224x224, Number of patches=14x142026.037695.498.164.590.895.8520.6
SGRBackbone=ViT, Resolution=384x384, Number of patches=24x242026.0375.895.798.665.69296.5524.2
VSE++Backbone=ViT, Resolution=224x224, Number of patches=14x142026.037594.69862.789.494.9514.6