Generic Video Instance Retrieval on This-Is-My personalization P (test)
56.4mAPMeta-Personalized Vision-Language Model
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Meta-Personalized Vision-Language Model2023.06 | 56.4 | 87.4 | |
| CLIP (V+L)Modality=visual + language2023.06 | 51.7 | 81.9 | |
| CLIP (visual)Modality=visual2023.06 | 48 | 75 | |
| CLIP (language)Modality=language2023.06 | 16.6 | 44.2 | |
| Random2023.06 | 1.1 | 2.2 |