BLIP support

[shalak]

trafficstars

Describe the feature you'd like to request

BLIP is a model trained to generate a caption based on the content of an image. Here are the examples of its work.

Describe the solution you'd like

In general, the captioning should be searchable by nextcloud's search feature.

I'm not sure what would be the best way to store the captioning (as a comment to a picture, as a separate .txt file to be created aside?)

Describe alternatives you've considered

I'm unaware of other models that work as good as BLIP.