Pyyan / Compare / Faster R-CNN vs Grounding DINO
Faster R-CNN vs Grounding DINO
2 of 5Object Detection · verified 13 Aug 2026
Faster R-CNNMicrosoftcurrentIGrounding DINOIDEA Researchcurrent
| Specification | Faster R-CNN | Grounding DINO |
|---|---|---|
| Summary | The two-stage baseline, still a reference point. | Detects whatever you describe in words. |
| COCO mAP | ~42 | ~52 zero-shot |
| Speed | Low | Moderate |
| Licence | MIT | Apache 2.0 |
| Family | Two stage | Open vocabulary |
| Open vocabulary | No | Yes |
| Released | 2015 | 2023 |
| Category | Object Detection | Object Detection |
| Official | Microsoft ↗ | IDEA Research ↗ |
Highlighted rows are where these differ.
Faster R-CNN
- Region proposals then refinement
- Included as the historical baseline
Best for accuracy where latency does not matter.
Grounding DINO
- Text prompt in, boxes out
- Fine-tuning still beats it on specialised objects
Best for classes you have no data for.