Submit a Project

ML x ART

A showcase with creative machine learning experiments

👋 I'm Emil, a resident at the Google Arts & Culture Lab. I've curated this collection.

Year
Artist
Technology
Theme
Medium
Default order

Google Arts & Culture Lab, Cyril Diagne

Take a selfie, extract the CNN features, and match them with features from portraits

CNN, NNS, Camera

Christian Mio Loclair, Waltz Binaire

A computer looks at itself with a camera and translates image features into text with an RNN

Camera, CNN, RNN

Janelle Shane, Google

An RNN that generates knitting instructions, then a knitter follows the instructions

RNN

Seung Joon Choi

Combining CPPN visuals and combining it with raymarching, VAE generated music

CPPN, VAE, Raymarching, ANN

Benjamin Wilkins, J.H. Gold

Use a CNN to classify low fidelity wireframes to mobile app components

CNN, Classification, Camera

Michael Kozlowski, Kishi Yuma

Display installation with GAN interpolations with photos of eyes with makeup

GAN

Merzmensch

A montage with several 3D Ken Burns Effect generated videos from images

3D Ken Burns Effect

Daniël Korssen

Create a shared playlist on Spotify and create a digital dance party by animating each dancer with PoseNet

PoseNet

Obvious

GAN trained on masks, and make the masks according to the GAN-generated images

GAN

Erik Swahn

GAN trained on Defra's Lidar data of London, then visualize the height maps in Aeriolod

StyleGAN, GAN, Aeriolod

Ross Goodwin, Oscar Sharp

An LSTM generated screenplay trained on screenplays from several films and television series, and produced according to the screenplay

RNN, LSTM

Alexander Mordvintsev

Run an image through a pre-trained CNN, and then upscale the image backward. For each layer, adjust the image according to the difference between the activations and the image

CNN, Activations, DeepDream, Gradient Ascent

Stanislas Chaillou

Using a GAN to generate architecture plans

GAN

Javier Ideami

Visualizing the loss landscapes with real data using ResNet 20 architectures with batch sizes of 16 and 128 and the Adam optimizer

Loss, ResNet, Adam

Travis LeRoy Southworth

Using the artist's own artworks to create a Style Transfer on portraits found online and animated in After Effects

Style Transfer

Sam Lavigne

Using semantic segmentation trained on Figaro-1k to extract all the haircuts of Mark Zuckerberg

Semantic segmentation, Figaro-1k

Jonas Jongejan, Henry Rowley, Takashi Kawashima, Jongmin Kim, Nick Fox-Gieg, Google Creative Lab

A user is challenged to draw an object and a CNN model tells if you are correct

CNN, Classification

Michael Kelly

GAN generated images of Edo woodblock art

GAN

Jessica In, George Profenza, Sam Price

A live, interactive drawing installation that facilitates collaboration between a human and a robot, the model is trained on the QuickDraw dataset and uses the QuickDraw Classifier, and a custom SketchRNN

Hardware, SketchRNN, QuickDraw, Classifier

Lulu xXX, Jacob Sutton

Apply Slic Superpixels and Style2Paint to Jacob Sutton's music video

Style Transfer

Robbie Barrat

Kanye West RNN lyrics generator

RNN

Sarah Meyohas

A GAN trained on a personal dataset of 100k individual petals and interpolated in a collage

GAN

Lin Fei Fei, Vadim Epstein

Several CV processing techniques have been applied to the paintings of Lin Fei Fei to turn them into an animated story

3D photo inpaint, First Order Motion, StyleGAN2, ConSinGAN, Video interpolation

Lulu xXX, Noj Barke

Use photogrammetry to extract vector points from images, and apply a Optical Flow based StyleTransfer with Noj Barke's dot paintings

Photogrammetry, Optical Flow, StyleTransfer, deepflow

Submit projects here 💅

☞ ML x ART on Twitter

Copyright © 2021 Emil Wallner

All artworks © of their respective owners