Video Conversation on Video-ChatGPT benchmark ActivityNet-200 (test)
3.08CorrectnessMiniGPT4-video
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MiniGPT4-videoLLM backbone=Llama 2-7B, Using Subtitles=true2024.04 | 3.08 | 3.02 | 3.57 | 2.65 | 2.67 | |
| MiniGPT4-videoLLM backbone=Mistral-7B, Using Subtitles=true2024.04 | 3.05 | 2.61 | 3.24 | 2.42 | 2.42 | |
| MiniGPT4-videoLLM backbone=Mistral-7B, Using Subtitles=false2024.04 | 2.97 | 2.58 | 3.17 | 2.38 | 2.44 | |
| LLaMA-VID-7BUsing Subtitles=false2024.04 | 2.96 | 3 | 3.53 | 2.46 | 2.51 | |
| MiniGPT4-videoLLM backbone=Llama 2-7B, Using Subtitles=false2024.04 | 2.93 | 2.97 | 3.45 | 2.47 | 2.6 | |
| BT-Adapter-7BUsing Subtitles=false2024.04 | 2.68 | 2.69 | 3.27 | 2.34 | 2.46 | |
| Video-ChatGPTUsing Subtitles=false2024.04 | 2.4 | 2.52 | 2.62 | 1.98 | 2.37 | |
| Video ChatUsing Subtitles=false2024.04 | 2.23 | 2.5 | 2.53 | 1.94 | 2.24 | |
| LLaMA AdapterUsing Subtitles=false2024.04 | 2.03 | 2.32 | 2.3 | 1.98 | 2.15 | |
| Video LLaMAUsing Subtitles=false2024.04 | 1.96 | 2.18 | 2.16 | 1.82 | 1.79 |