Referring Expression Comprehension on gRefCOCO (testB)
54.8Pr(F1=1, IoU>=0.5)SimVG-DB
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SimVG-DBVisual Encoder=ViT-B/32, Textual Encoder=N/A2024.09 | 54.8 | 57.2 | |
| SimVG-TBVisual Encoder=ViT-B/32, Textual Encoder=N/A2024.09 | 53.1 | 57.5 | |
| UNINEXTVisual Encoder=RN50, Textual Encoder=BERT2024.09 | 42.9 | 48.2 | |
| MDETRVisual Encoder=RN101, Textual Encoder=RoBERTa2024.09 | 36.5 | 31 | |
| MM-G-T(c3)Backbone=Swin-T, Setting=zero-shot2024.01 | 31.6 | 96.7 | |
| LLMDetBackbone=Swin-T, Zero-shot=true2025.01 | 30.5 | 92.9 | |
| MM-GDINOBackbone=Swin-T, Zero-shot=true2025.01 | 30.4 | 92.3 | |
| VLTVisual Encoder=DN53, Textual Encoder=GRU2024.09 | 30.2 | 32.5 | |
| G-DINO-T(c)Backbone=Swin-T, Setting=zero-shot2024.01 | 30 | 86.1 | |
| Grounding-DINOBackbone=Swin-T, Zero-shot=true2025.01 | 30 | 86.1 | |
| MCNVisual Encoder=DN53, Textual Encoder=GRU2024.09 | 26.8 | 30.3 |