Generative 3D Modelling for Game Asset Creation using Deep Learning Techniques

The 3D entertainment industry has expanded quickly in recent years, largely driven by animated content, streaming services, video game development, AR/VR/XR. The new trend enabled by the ubiquitous graphics processing power is that users are becoming creators. Surprisingly, the fundamentals of 3D creation have not changed in 45 years. This puts the creation of 3D out of reach for 99.8% of consumers and, for professionals, state-of-the-art methods are still too costly and labor intensive to practically meet growing 3D demand. Generative Adversarial Networks (GANs) and other AI technologies offer new possibilities in the AI generation of 3D and 2D art assets. This project investigates the potential of using deep learning techniques for generative 3D modelling for game development and has significant implications for the industry revenue stream.

Faculty Supervisor:

Ali Mahdavi-Amiri

Student:

Partner:

Tori Technologies

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Simon Fraser University

Program: