Audio-referred visual grounding on AVISeg (test)
18.55FSLAAL-Ref-SAM 2
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| AL-Ref-SAM 2Assistant=GPT-4, Zero-shot=true2023.10 | 18.55 | 38.02 | 15.84 | |
| PG-Video-LLaVA-7BAssistant=GPT-3.5, Zero-shot=true2023.10 | 9.15 | 22.86 | 5.94 | |
| BuboGPT-7BAssistant=GPT-4, Zero-shot=true2023.10 | 7.75 | 20.16 | 5.76 | |
| Sam4AVSZero-shot=true2023.10 | 0 | 8.18 | 3.93 |