An experiment using #pixel2style2pixel to create real people from drawings. The AI applies the face shape to a random person in the model.
Source video by @bobbo_andonova. Big thanks for sharing the great drawings! 🙏
More examples + high quality video link in thread. 📽️ ⬇️
Source video by @bobbo_andonova. Big thanks for sharing the great drawings! 🙏
More examples + high quality video link in thread. 📽️ ⬇️
This video uses a @Procreate timelapse from character designer Randy Bishop as input.
The #pixel2style2pixel model often sees eyes as glasses. It's like a #StyleGAN version of #DeepFaceDrawing @SpirosMargaris @pascal_bornet
Randy's IG: instagram.com/randybishopart/ Thanks Randy!
The #pixel2style2pixel model often sees eyes as glasses. It's like a #StyleGAN version of #DeepFaceDrawing @SpirosMargaris @pascal_bornet
Randy's IG: instagram.com/randybishopart/ Thanks Randy!
And here's what the neural network does crappy drawing as it's input! Future of criminal sketch art?
Any #Procreate character illustrators want to collab on one of these? I've got ideas!
Links to HQ versions of these videos on my site here: nathanshipley.com/gan 📽️
Any #Procreate character illustrators want to collab on one of these? I've got ideas!
Links to HQ versions of these videos on my site here: nathanshipley.com/gan 📽️
I also owe big thanks to @_ScottEaton_ and @memotv for the inspiration to play with ML/GAN in the first place last year.
Scott's deep draw lecture and Memo's Learning to See were formative for me and still are!
Just shared both of your work yesterday in a talk at @UMArtDept 🙏
Scott's deep draw lecture and Memo's Learning to See were formative for me and still are!
Just shared both of your work yesterday in a talk at @UMArtDept 🙏
• • •
Missing some Tweet in this thread? You can try to
force a refresh
