md-mohaiminul/VideoRecap

Usage of Pre-Trained Models

Closed this issue · 2 comments

Hello,
Apologies for my ignorance, but how can I use the pre-trained models to caption a video? I see instructions for extracting features, training, and evaluating the model, but I can't figure out how to generate captions.
Thank you!

Thanks for your comments. We have added a demo notebook (demo.ipynb) to extract three-levels of hierarchical captions from any video using our pretrained models.

Cool, thanks so much! Must have missed that, sorry about that!