Thanks to @matthen2, I got to test out my new candidate DeOldify model on a use case that I think will be quite appealing for many people: wedding photos! Thanks Matt!pic.twitter.com/OEscz5XhiW
Show this thread
-
Amazing work
@citnaj. Simpleton Question: I was wondering — would it make sense to have a domain-classifier at the head end of the process to decide which model to use? Eg “Oh this looks like a portrait, a wedding portrait, so let me use the wedding portrait trained colorizer.”
Think about the alternative- how many models would you wind up having to satisfy all scenarios if you had multiple models?
-
-
No question, many, many. I suppose this happens as well in, say, NLP where inbound language could be any language.
-
Yes that’s actually a great example where training in many languages benefits performance in all said languages in a mutually beneficial way. More correct generalizations can be modeled, I gather.
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.