Interesting Machine Learning Projects
Image translation without input-output pairs.
Give computer two pictures and it will redraw the first one with the style of the second one.
The original Style Transfer requires lots of computational resources, this is simpler but faster version.
From left to right, the first column is the 16x16 input image, the second one is what you would get from a standard bicubic interpolation, the third is the output generated by the neural net, and on the right is the ground truth.
Computer restores the missing part of the picture.
Tell computer how flower should looks like and it will draw it.
Computer creating little stories about pictures.
Image Super-Resolution for Anime-Style Art.
Turn your two-bit doodles into fine artworks with deep neural networks, generate seamless textures from photos, transfer style from one image to another, perform example-based upscaling, but wait... there's more! (An implementation of Semantic Style Transfer.) Video.
Computer tells if selfie is good or bad. Sorted selfies on the picture below - good at the top bad at the bottom.
There's also similar tool from Google.
There's also another project doing the same
And, similar project with the code and explanation.
Makes people smile, or removes smile.