Follow

I'm playing with U-Nets. You get some pretty interesting results when you pass an image into a model multiple times.

@Wetrix it's the result of a machine learning app I'm working on. I trained a U-Net(a machine learning model) to tilt eyebrows up, then I ran a picture of me through it 20 times. What you see is what happened when I did that.

I'm working on creating a GAN so I can get photorealistic results, but it isn't working yet.

@lhackworth Photos can get super-weird using multi-passes thru neural nets. Have you ever seen DeepDream? youtube.com/watch?v=SCE-QeDfXt

@Pat amazing! I'd seen it years ago, but that was well before I knew much about how a neural net worked. I'm gonna have to make a crappy clone of that one day

@lhackworth I don't know anything about neural nets, but your thumbnail on ML is perfect -- instead of the usual process of supplying an input to a known function to produce an output, you flip it around and get machine to produce the function. Cool.

Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves. A STEM-oriented instance.

An inclusive free speech instance.
All cultures and opinions welcome.
Explicit hate speech and harassment strictly forbidden.
We federate with all servers: we don't block any servers.