4D Scanning and Automated Reconstruction of Digital Doubles for Visual Effects

Modern visual effect films rely on photorealistic, believable, synthesized humans and characters. Often these characters are used as stand-ins for the hero character in a film and are referred to as digital doubles. In the past, digital doubles have been limited in use due to cost or complexity. As the demand for the use of digital doubles in more aspects of modern filmmaking rises, we seek solutions to accelerate and automate the creation of digital doubles. This project will explore approaches to capture sufficient high-quality details of an actor’s performance, in neutral or synthesized lighting conditions, to generate a synthetic human.

Faculty Supervisor:

Eitan Grinspun;David Lindell

Student:

Partner:

DNEG

Discipline:

Computer science

Sector:

Information and cultural industries

University:

University of Toronto

Program: