facebookresearch / LaViLa

Code release for "Learning Video Representations from Large Language Models"
MIT License
478 stars 42 forks source link

Extracting spatial feature maps from LaViLa #27

Open vineetparikh opened 11 months ago

vineetparikh commented 11 months ago

Hi there, thanks so much for the cool work and docs!

I'm looking to use the TSF-B video encoder on EK100 data, and want to extract spatial feature maps from LaViLa for each frame. Are there any ways to do this/any code to do this?