Open-vocabulary Attribute Detection on OVAD
29.2mAP (All)X2-VLMlarge
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| X2-VLMlargeMode=Zero-shot, Backbone=Large2022.11 | 29.2 | 55.1 | 36.4 | 11.3 | |
| X2-VLMlarge4MMode=Zero-shot, Backbone=Large, Pre-training Data=4M2022.11 | 27.7 | 54 | 34.4 | 10.1 | |
| X2-VLMbaseMode=Zero-shot, Backbone=Base2022.11 | 27.6 | 52.2 | 34.7 | 10.3 | |
| BLIPMode=Zero-shot2022.11 | 24.3 | 51 | 28.5 | 9.7 | |
| X2-VLMbase4MMode=Zero-shot, Backbone=Base, Pre-training Data=4M2022.11 | 24 | 51.9 | 29.7 | 7.2 | |
| OVADetectorMode=Zero-shot2022.11 | 21.4 | 48 | 26.9 | 5.2 | |
| CLIPMode=Zero-shot2022.11 | 17 | 44.3 | 18.4 | 5.5 | |
| ALBEF4MMode=Zero-shot, Pre-training Data=4M2022.11 | 15.6 | 43.1 | 17.3 | 3.7 | |
| RandomMode=Zero-shot2022.11 | 8.6 | 36 | 7.3 | 0.6 |