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

Justin Pinkney

Making Ukiyo-e portraits real using resolution-dependent model interpolation for StyleGAN

GAN, StyleGAN

Cyril Diagne

Use BASNet to extract an object from the environment and OpenCV SIFT to add it into photoshop

BASNet, OpenCV, SIFT, Camera

Michael Kelly

GAN generated images of Edo woodblock art

GAN

Erwin Driessens, Maria Verstappen

Using Face Recognition and a robot arm to find grains of sand that looks like faces

Face Recognition, CNN

Nikita Pekin

Text-to-image for "A painting of a murder in the style of Monet" generated by The Big Sleep

Big Sleep, CLIP

Christian Mio Loclair, Waltz Binaire

Adjust user interface animations according to an ANN trained on user interactions

ANN

Hans Brouwer

Audio-reactive latent interpolation with StyleGAN2 created by modulating latent and noise inputs with musical features

StyleGAN2

Zaria Howard

Draw a face in VR and translate it to a real face with a Conditional GAN

pix2pix, VR, GAN

Ben Snell

A 3D-GAN trained on 1000 sculptures inspired by the Greek god Dionysus

GAN, 3D

DataDada

Train a model to map a gesture to a sound

PoseNet

Erik Swahn

Four-channel StyleGAN trained on SRTM elevation data

StyleGAN, GAN

Alexandre Devaux

Using BodyPix to segment people and WebXR to create holograms on-the-fly

BodyPix, Tensorflow.js

Gene Kogan

Continuously zoom in a image and apply DeepDream and use Dense Captioning to label objects in the video

DenseCap, DeepDream

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

Sergii Kharagorgiiev

Generative art program synthesis using Three.js, a general purpose graphics framework

Three.js

Jürgen Schmidhuber, Douglas Eck

Blues Improvisation with an LSTM

RNN, LSTM

Alsino Skowronnek

A StyleGAN trained on thousands of frames of a filmed sequence of images of graffiti from NYC in the late 70s/early 80s. The output is interpreted into a painting and an AR filter

StyleGAN

Yotam Mann, Google Creative Lab

A tool to mix the latent space of two instruments

VAE, Magenta

Andy Matuschak, Google

A real time implementation of Sketch RNN which predicts future strokes while you draw

RNN, QuickDraw

Magenta

Translating everyday sounds into instruments with Magenta's Differentiable Digital Signal Processing

DDSP, Magenta

Joel Simon

Use a genetic algorithm to generate faces and the dlib face classifier for the objective function

Face Recognition, CNN

Stefan Sietzen

Realtime Interactive Visualization of Convolutional Neural Networks in Unity

CNN, Unity

Alsino Skowronnek

A Conditional GAN trained on a data set of ~1,000 annotated street tag images from Germany. The model can generate new graffiti tags based on any textual input from the audience

Pix2Pix

Alexander Mordvintsev

Use Style Transfer where the content comes from a face and the style from a video of clouds

Style Transfer

Submit projects here 💅

☞ ML x ART on Twitter

Copyright © 2021 Emil Wallner

All artworks © of their respective owners