In the last few years, generative machine learning (ML) models have been rapidly improving and changing the landscape for creating images. Abstract concepts like “style” and “content” can now be expressed through statistics via algorithms. This is now possible due to the advances that have been made in the processing power of GPUs (graphics cards).
In addition to this, Google has made access to GPU resources and Python notebook environments available on the web for free through Google Colaboratory. This makes it possible for anyone with no coding or technical background to use machine learning models for creative exploration. In this workshop, we will learn how to use these models for creating your own custom Snapchat filters.
In this workshop, participants will:
- Learn the basics of Google Colab and using premade notebooks.
- Learn how to train a custom style transfer model that can be used as a Snapchat lens/filter using a premade Google Colab notebook.
- Learn the basics of Lens Studio, Snapchat’s tool for user created lenses.
- Deploy your lens (if you’d like) and share it with your friends in only a few hours!
This is an introductory class, so no previous experience with coding, premade notebooks, or Google Colab is required! It would be helpful to have had some experience using Snapchat, and have a Snapchat account.
About the workshop instructor: Miles Vesh Miller is a digital artist and Machine Learning Engineer at Snap. He currently resides in San Francisco.