Image-to-Text Retrieval on CamoIT 3K (test)
15.1R@1CUSA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CUSATrained on=MS-COCO, Pub.=24, FG=false, BU=false2026.04 | 15.1 | 37 | |
| D2S-VSETrained on=MS-COCO, Pub.=25, FG=false, BU=false2026.04 | 13.3 | 33.1 | |
| CUSATrained on=Flickr30K, Pub.=24, FG=false, BU=false2026.04 | 12.6 | 34 | |
| LAPSTrained on=MS-COCO, Pub.=24, FG=true, BU=false2026.04 | 11.8 | 29.8 | |
| HREMTrained on=MS-COCO, Pub.=23, FG=false, BU=true2026.04 | 11.3 | 28.1 | |
| CHANTrained on=MS-COCO, Pub.=23, FG=true, BU=true2026.04 | 10.9 | 29.7 | |
| CFMTrained on=MS-COCO, Pub.=22, FG=false, BU=true2026.04 | 10.7 | 26 | |
| D2S-VSETrained on=Flickr30K, Pub.=25, FG=false, BU=false2026.04 | 7.5 | 20.9 | |
| DBLTrained on=MS-COCO, Pub.=24, FG=true, BU=true2026.04 | 7.1 | 20 | |
| CHANTrained on=Flickr30K, Pub.=23, FG=true, BU=true2026.04 | 6.9 | 20 | |
| HREMTrained on=Flickr30K, Pub.=23, FG=false, BU=true2026.04 | 6.7 | 19.2 | |
| LAPSTrained on=Flickr30K, Pub.=24, FG=true, BU=false2026.04 | 5.9 | 17.2 | |
| AVSETrained on=Flickr30K, Pub.=25, FG=false, BU=false2026.04 | 5.7 | 16 | |
| CFMTrained on=Flickr30K, Pub.=22, FG=false, BU=true2026.04 | 5.4 | 16.5 | |
| DBLTrained on=Flickr30K, Pub.=24, FG=true, BU=true2026.04 | 4.5 | 14.2 |