OpenBMB / MiniCPM-V

MiniCPM-Llama3-V 2.5: A GPT-4V Level Multimodal LLM on Your Phone
Apache License 2.0
7.82k stars 543 forks source link

对于视频数据,除了抽帧再单独处理每一张之外,是否有更好的方法呢? #317

Open xyw1 opened 2 days ago

xyw1 commented 2 days ago

是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?

该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?

当前行为 | Current Behavior

需要根据视频总结出摘要,视频拍摄角度有时候很混乱,先抽帧生成文本再总结的效果不好

期望行为 | Expected Behavior

No response

复现方法 | Steps To Reproduce

No response

运行环境 | Environment

- OS:
- Python:
- Transformers:
- PyTorch:
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):

备注 | Anything else?

No response

LDLINGLINGLING commented 17 hours ago

目前我们的模型只有输入单张图片的能力,因此还是要处理成图片,好像没能想到更好的办法。