Image-to-text retrieval on ECCV Caption (test)
84.9R@1VACSR
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VACSRBackbone=CLIP ViT-B/162026.05 | 84.9 | 45.5 | 35.4 | |
| NeighborRetr2025.03 | 82.1 | 45.6 | 34.4 | |
| PCME++Backbone=CLIP ViT-B/162026.05 | 81.6 | 45.1 | 34.5 | |
| PCME++2025.03 | 81.3 | 45 | 34.4 | |
| CUSA2025.03 | 80.9 | 44.1 | 33.6 | |
| VACSRBackbone=CLIP ViT-B/322026.05 | 80.6 | 43.8 | 33.2 | |
| PCMEBackbone=CLIP ViT-B/162026.05 | 79.1 | 44 | 33.2 | |
| PCME++Backbone=CLIP ViT-B/322026.05 | 76.6 | 43.4 | 32.3 | |
| DAABackbone=CLIP ViT-B/322026.05 | 75.9 | 42.3 | 31.2 | |
| PCMEBackbone=CLIP ViT-B/322026.05 | 74.9 | 42.3 | 31.2 | |
| ViLT2025.03 | 73.4 | 38.6 | 27.5 | |
| P2RMBackbone=CLIP ViT-B/162026.05 | 72.9 | 42.2 | 30.6 | |
| P2RMBackbone=CLIP ViT-B/322026.05 | 72.2 | 41.7 | 30.2 | |
| SGRAF2025.03 | 71.1 | 39.3 | 27.4 | |
| PCME2025.03 | 65.5 | 38.7 | 26.2 | |
| PVSE2025.03 | 62.6 | 35.6 | 23.4 | |
| DAABackbone=CLIP ViT-B/162026.05 | 40.3 | 22.4 | 12.4 |