Environment Design for Virtual Production in Unreal Engine 4

Virtual production is an emerging filmmaking method that uses real-time rendering to combine live-action footage with CG, first used on films like The Lion King and The Mandalorian TV series. In this 2-hour workshop, Vladimir Somov, former Principal Cinematics Artist on Assassin’s Creed, For Honor, and Far Cry, shares the core concepts and techniques required for creating Virtual Production-ready environment art using Unreal Engine. He walks through the entire pipeline of creating the exterior and interior sets of an ancient temple — an environment that allows for the exploration of two scenarios for lighting and mood, so you can learn how to set up and tell a story with lighting and post-processing for different environments.

The workshop begins with a lesson on understanding how to work with references and shows how to create a custom construction kit and a trim material. You will learn how to set up Unreal Engine, what project settings and plugins to use and why, as well as how to organize and structure Unreal projects. Vladimir provides both practical and theoretical advice on effective and believable set dressing and lighting, and offers techniques to profile and optimize performance with tools like Deep Learning Super Sampling (DLSS) from Nvidia. The workflow also includes a look inside the techniques used for capturing a basic cinematic with an animated character and a virtual hand-held camera operated by using a smartphone as a tracker.

While this workshop is targeted at beginners, some basic knowledge and understanding of Unreal Engine and 3D modeling is recommended before starting. 3ds Max, Quixel Bridge, and Quixel Mixer are used in conjunction with Unreal Engine 4. At the end of 2 hours, you should have a strong understanding of the entire process involved in creating real-time environment art, and feel confident with techniques that can be scaled up and applied to anything from a full LED volume Virtual Production to Cinematics and Post-Visualization.