search

keplerlab / katna

Tool for automating common video key-frame extraction, video compression and Image Auto-crop/Image-resize tasks
https://katna.readthedocs.io/
MIT License
313 stars 59 forks source link

train? #15

Closed johnbager closed 3 years ago

johnbager commented 3 years ago

is it can be trained in my dataset?

mayank10j commented 3 years ago

No, Most of the modules do not need training, for some of the processes like Text detection, face detection we use pretrained models from opencv. No plans for adding training support for them as of yet.