An artificial neural network making predictions on live webcam input, trying to make sense of what it sees, in context of what it’s seen before.
It can see only what it already knows, just like us.
(Trained on tens of thousands of images scraped from flickr. not ‘style transfer’ 🙂
http://www.memo.tv/learning-to-see-you-are-what-you-see/
Music: Diamanda Galas – ‘Gloomy Sunday’
code based on (but more evolved version of)
https://github.com/memo/webcam-pix2pix-tensorflow
model (ie training + inference) based on
https://github.com/affinelayer/pix2pix-tensorflow
In turn based on
https://phillipi.github.io/pix2pix/
https://arxiv.org/abs/1611.07004
In turn based on
https://arxiv.org/abs/1406.2661
and
https://github.com/Newmu/dcgan_code
https://arxiv.org/abs/1511.06434
In turn based on
https://github.com/goodfeli/adversarial
https://arxiv.org/abs/1406.2661
In turn based on
http://people.idsia.ch/~juergen/deep-learning-overview.html