Playful Exploration of U-Net for Glitch Effects

"UGly-Net" is a playful interactive tool that invites users to tinker with and create their own video synthesis U-Nets. In doing so, users can learn by changing the feature maps of the model and experiment on their creative expression using deep neural networks.

UGly-Net introduces the U-Net structure in simplified visualization with bold strokes and vivid colors. It is intended to offer users a playful, humorous, and creative take on deep neural networks.

Users can start from a well-trained and stable U-Net for slow-motion video synthesis, and then transform the network by freely making modifications to each part of the network. Users can dictate how to "carve-out" the inputs of each layer, resulting in unique U-Nets for each user. UGly-Net changes the network feature maps according to user-generated variations to synthesize videos with glitchy effects. The videos are then collected and displayed in the gallery along with its corresponding U-Net. A user's creative personality is expressed through both the unique visual output of U-Net itself and its video output.

Changing each layer's feature maps of the encoder and decoder produced a wide variety of video effects. Some resembled classical glitch effects on cathode-ray tube televisions or 8-bit computers. Others produced genuine and inspiring patterns that are rarely seen in previous sets of commercial glitch effects or artworks.

Artificial intelligence (AI) is increasingly distancing itself from the general public with its rising complexity. Although AI possesses rich potential as a creative medium, it is seen by many people as intimidating rather than inspiring.

In the process of "doughing" the U-Net iteratively and synthesizing videos, the user instinctively learns the role of each part of the U-Net and its effects on videos. By creating glitch effects with UGly-Net, users can manipulate the network on their own and realize its limitations and expressiveness. While previous digital generations manipulated retro computers and formed their own understanding of computers, UGly-Net can help users understand the new "untrodden realm" of neural networks.

UGly-Net is also an artistic means of expressing individuality. Contemporary computers deliver flawless images that are nearly uniform in quality across operating systems and hardware. At this time, the glitch created by the user is a unique creative expression that only the user's own video and set of vectors between layers can produce.

The slow iterative work is both an artistic practice and a process of relating to technical concepts.

Stills & Videos


Seri Lee

Pulse9 Inc.

Interested in machine learning. #Image #Video #VAE #Style Transfer #Developer

Eun Jee Sung


A human-ai interaction enthusiast. HAI, machine learning, and media arts.

Juhee Lee


One of the main interests in my research is algorithm of intelligent Control. Currently, I conduct a research project on the creativity A.I. for deep learning including mathematics and cryptography for machine learning.

Jieun Park

Pulse9 Inc.

Founder of Pulse9 Inc., where we have built PaintlyFX, an Creative AI Platform that enables everyone to easily make their own artwork without professional skills.