Embodied Dialog Localization on WAY Seen (val)
25.64Accuracy (0)BLIP(DiaLoc)†
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BLIP(DiaLoc)†Mode=Single-shot2024.03 | 25.64 | 66.02 | |
| LingUNetMode=Single-shot2024.03 | 19.87 | 59.29 | |
| DiaLoc-iMode=Multi-shot2024.03 | 18.43 | 57.18 | |
| DiaLoc-eMode=Multi-shot2024.03 | 18.36 | 60 | |
| LingUNet-eMode=Multi-shot2024.03 | 14.47 | 46.15 | |
| BLIP-Res18Mode=Single-shot2024.03 | 13.44 | 50.82 | |
| LingUNet-iMode=Multi-shot2024.03 | 4.32 | 24.16 |