Video-based text generation on ActivityNet-200
2.44Correctness of InformationVISTA-LLAMA
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VISTA-LLAMA2023.12 | 2.44 | 2.31 | 2.64 | 3.18 | 2.26 | |
| Video-ChatGPT2023.12 | 2.4 | 2.37 | 2.52 | 2.62 | 1.98 | |
| Video Chat2023.12 | 2.23 | 2.24 | 2.5 | 2.53 | 1.94 | |
| LLaMA Adapter2023.12 | 2.03 | 2.15 | 2.32 | 2.3 | 1.98 | |
| Video LLaMA2023.12 | 1.96 | 1.79 | 2.18 | 2.16 | 1.82 |