Image-to-Text Retrieval on RSICD (test) with RSUM
18.12R@1ELC
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ELCBackbone=ViT-B/322026.07 | 18.12 | 38.55 | 51.78 | 215.2 | |
| ELC0Backbone=ViT-B/322026.07 | 17.93 | 37.88 | 51.88 | 213.47 | |
| RemoteCLIPBackbone=ViT-B/322026.07 | 17.02 | 37.97 | 51.51 | 211.57 | |
| FGVLABackbone=ViT-B/322026.07 | 16.93 | 36.23 | 48.49 | 204.74 | |
| AIRBackbone=ViT-B/322026.07 | 16.29 | 33.58 | 46.39 | 199.46 | |
| VSE∞Backbone=ViT-B/322026.07 | 16.1 | 36.51 | 48.76 | 193.08 | |
| VSE++Backbone=ViT-B/322026.07 | 16.02 | 35.04 | 47.58 | 185.35 | |
| PolyBackbone=ViT-B/322026.07 | 15.55 | 32.02 | 45.2 | 181.18 | |
| InfoNCEBackbone=ViT-B/322026.07 | 15.28 | 33.39 | 47.48 | 196.49 | |
| PAUBackbone=ViT-B/322026.07 | 15.1 | 34.31 | 48.12 | 198.39 | |
| GLGSABackbone=ViT-B/322026.07 | 14.38 | 36.73 | 45.35 | 192.4 | |
| PE-RSITRBackbone=ViT-B/322026.07 | 14.13 | 31.51 | 44.78 | 186.7 | |
| CUPBackbone=ViT-B/322026.07 | 12.05 | 32.66 | 48.7 | 182.02 | |
| KAMCL2026.07 | 11.99 | 27.17 | 38.33 | 150.52 | |
| SIRSBackbone=ViT-B/322026.07 | 10.34 | 27.69 | 39.66 | 166.31 | |
| SMLGN2026.07 | 8.87 | 25.53 | 37.24 | 149.21 | |
| GaLR2026.07 | 6.59 | 19.85 | 31.04 | 113.78 | |
| AMFMN2026.07 | 5.39 | 15.08 | 23.4 | 98.49 | |
| CLIPBackbone=ViT-B/322026.07 | 5.03 | 14.64 | 23.51 | 93.96 |