I went down a rabbit hole tonight running black and white art through my latest unreleased DeOldify model tonight. A brand new model as of this week. Enjoy!pic.twitter.com/wi6dwZwRLf
Show this thread
-
-
Image sources: https://commons.wikimedia.org/wiki/File:Moritz_Daniel_Oppenheim_-_Purim_(Feast_of_Esther)_(Das_Purim-Fest)_-_Google_Art_Project.jpg … https://commons.wikimedia.org/wiki/File:Moritz_Daniel_Oppenheim_-_The_Wedding_(Die_Trauung)_-_Google_Art_Project.jpg … https://commons.wikimedia.org/wiki/File:A_young_woman_watching_from_a_beach_while_a_rowing_boat_carr_Wellcome_V0038708.jpg … https://commons.wikimedia.org/wiki/File:A_dishevelled_Tom_Rakewell_in_the_gambling_house,_the_obsess_Wellcome_V0049291.jpg … https://commons.wikimedia.org/wiki/File:Liverpool_Food_and_Betterment_Association._Wellcome_L0000899.jpg … https://commons.wikimedia.org/wiki/File:R_Staines_Malvolio_Shakespeare_Twelfth_Night.jpg … https://commons.wikimedia.org/wiki/File:Eug%C3%A8ne_Carri%C3%A8re_-_Still_Life_-_1967.183_-_Art_Institute_of_Chicago.jpg … https://commons.wikimedia.org/wiki/File:Luigi_Loir_-_The_Boathouse_-_1962.816_-_Art_Institute_of_Chicago.jpg …
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How large is the dataset you’re training deoldify on? Thinking of maybe trying danbooru2018....
-
About 2.5M images, so pretty comparable. I think that's a great idea to try training on large scale anime- lots of potential and from what I understand a lot of people will be happy about it (I've heard requests/queries multiple times about it).
- 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.