Hacker News new | ask | show | jobs
by remixer-dec 1036 days ago
If I understand correctly, you have to train a custom NLA model for each video segment before actually using it.