search

boheumd / MA-LMM

(2024CVPR) MA-LMM: Memory-Augmented Large Multimodal Model for Long-Term Video Understanding
https://boheumd.github.io/MA-LMM/
MIT License
221 stars 26 forks source link

Can you provide the code for video data preprocessing? #17

Closed longmalongma closed 3 months ago

longmalongma commented 4 months ago

We are currently following your masterpiece MA-LLM, but we are encountering issues with video data preprocessing. It seems that you do not have publicly available code for video data preprocessing. Can you provide the code for video data preprocessing? If possible, thank you very much.

boheumd commented 4 months ago

Hello, for the video data processing, you can refer to this issue https://github.com/boheumd/MA-LMM/issues/3 for more help. Currently, we extract frames from videos with 10 fps.

boheumd commented 3 months ago

Hi, I updated the README.md. The example preprocessing code is provided here https://github.com/boheumd/MA-LMM/blob/main/data/extract_frames.py