Visual Grounding on RefCOCOg umd (val)
80P@1APE (D)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| APE (D)Backbone=ViT-L2023.12 | 80 | 64.2 | — | |
| APE (B)Backbone=ViT-L2023.12 | 79.9 | 62.8 | — | |
| APE (C)Backbone=ViT-L2023.12 | 79.8 | 63.2 | — | |
| Grounding DINOBackbone=Swin-T2023.12 | 71 | — | — | |
| GLIPBackbone=Swin-T2023.12 | 66 | — | — | |
| KOSMOS-2Backbone=ViT-L2023.12 | 60.5 | — | — | |
| MDETRBackbone=ENB52023.12 | 57.1 | — | — | |
| APE (A)Backbone=ViT-L2023.12 | 38.9 | 28.1 | — | |
| QueryNLT2024.03 | — | — | 75.3 | |
| ReSC2024.03 | — | — | 67.3 | |
| TransVG2024.03 | — | — | 68.67 | |
| VLTVG2024.03 | — | — | 76 |