Referring Object Detection on RefCOCO (val)
89.57Top-1 AccuracyHeROD-G
Evaluation Results
| Method | Links | |
|---|---|---|
| HeROD-GFine-tuning=w/2026.03 | 89.57 | |
| Grounding DINO*Fine-tuning=w/2026.03 | 88.86 | |
| DQ-DETRFine-tuning=w/2026.03 | 88.63 | |
| MDETRFine-tuning=w/2026.03 | 86.75 | |
| RefTRFine-tuning=w/2026.03 | 85.65 | |
| VILLA-LFine-tuning=w/2026.03 | 82.39 | |
| HeROD-UFine-tuning=w/2026.03 | 81.93 | |
| TransVGFine-tuning=w/2026.03 | 81.02 | |
| VGTRFine-tuning=w/2026.03 | 79.2 | |
| UNINEXT*Fine-tuning=w/2026.03 | 78.61 | |
| HeROD-GSupport Data=2k, Stage=Support Set Training2026.03 | 78.17 | |
| HeROD-GSupport Data=2k, Stage=2k Finetune2026.03 | 78.17 | |
| HeROD-GSupport Data=2k, Stage=1k Finetune2026.03 | 76.9 | |
| HeROD-GSupport Data=1k, Stage=Support Set Training2026.03 | 76.81 | |
| HeROD-GSupport Data=1k, Stage=2k Finetune2026.03 | 76.81 | |
| MAttNetFine-tuning=w/2026.03 | 76.65 | |
| HeROD-GSupport Data=1k, Stage=1k Finetune2026.03 | 75.56 | |
| HeROD-GSupport Data=2k, Stage=0.5k Finetune2026.03 | 75.08 | |
| HeROD-GSupport Data=1k, Stage=0.5k Finetune2026.03 | 72.85 | |
| Grounding DINOSupport Data=2k, Stage=2k Finetune2026.03 | 67.23 | |
| Grounding DINOSupport Data=1k, Stage=2k Finetune2026.03 | 64.74 | |
| Grounding DINOSupport Data=2k, Stage=1k Finetune2026.03 | 62.98 | |
| Grounding DINOSupport Data=2k, Stage=0.5k Finetune2026.03 | 62.08 | |
| Grounding DINOSupport Data=1k, Stage=1k Finetune2026.03 | 61.81 | |
| Grounding DINOSupport Data=2k, Stage=Support Set Training2026.03 | 61.18 | |
| Grounding DINOSupport Data=1k, Stage=0.5k Finetune2026.03 | 60.65 | |
| Grounding DINOSupport Data=1k, Stage=Support Set Training2026.03 | 59.28 | |
| Grounding DINO*Fine-tuning=w/o2026.03 | 50.61 |