Genmo (short for Generative Mosaic) is an AI-powered app that recreates any photo or video using an entirely separate set of images. No original video content remains after processing - and no filters - the new clip is completely synthesized by our neural network. Take a picture or video, select a dataset to render it with, and get your recreated piece back in seconds!
We use our own deep neural network architecture, which hybridizes a generative adversarial network (GAN) with other neural network models. It's fast (relatively!) and, unlike other popular AI-based image manipulation tools, we do not use a pre-trained classification network. Image and video is redrawn at near real-time speed from an arbitrarily chosen image set on-demand.
Genmo is a project of LatentCulture, an AI research & production venture based within Jason Salavon Studio on the University of Chicago campus with support from the Neubauer Collegium. We are artists, programmers, and researchers investigating how deep learning and other generative systems can impact creative production. We welcome opportunities & partnerships to collaborate in this endeavor. Contact us at firstname.lastname@example.org.