CS 8803 O27: Computer Graphics in the AI Era
Instructional Team
Bo Zhu
Creator, Instructor
Overview
The CGAI course offers a comprehensive introduction to modern computer graphics, focusing on techniques that have emerged over the past decade, with a particular emphasis on AI-powered advancements in modeling, rendering, simulation, and animation. Central to the course are implicit neural representations, differentiable rendering, Gaussian splatting, neural physics simulation, and generative models, which have jointly revolutionized the field of computer graphics by enabling realistic, dynamic, and intelligent graphics systems.
The CGAI curriculum explores the cross-cutting intersections across multiple domains in contemporary AI-driven graphics applications, including interactive graphical systems, visual content creation pipelines, and differentiable rendering/animation frameworks. The core topics include Neural Signed Distance Fields (SDF), Neural Radiance Fields (NeRF), 3D Gaussian Splatting (3D-GS), differentiable and neural physics, generative models, and world models. These elements are further expanded with advanced topics, including generative 3D graphics pipelines using diffusion models and agentic frameworks. The course also connects the topics to traditional graphics realms such as shape representation, volumetric media, splatting, and physically-based animation.
Course Goals
After taking this course, students will develop a forward-looking understanding of computer graphics, gaining both theoretical insights and practical skills to create interactive and realistic content powered by AI-enabled techniques. Positioned as the next-level advanced computer graphics course following the introductory computer graphics, this course equips students with cutting-edge tools and a modern graphics mindset, preparing them to engage with the evolving landscape of AI-integrated graphics applications.
Sample Syllabus
Spring 2026 syllabus (PDF)
Note: Sample syllabi are provided for informational purposes only. For the most up-to-date information, consult the official course documentation.
Before Taking This Class...
Suggested Background Knowledge
Students are expected to be familiar with the fundamentals of computer graphics at the level of CS 3451, including basic concepts in geometry, transformations, rendering, and animation. For students without prior graphics background, the course begins with three preparatory lectures that provide a concise but sufficient overview of these foundations to ensure everyone starts from a common baseline.
In addition, a working understanding of calculus (differentiation and integration) and linear algebra (vectors, matrices, eigenvalues, etc.) is required.
No prior background in AI/ML is required.
Technical Requirements and Software
Students should have access to a laptop equipped with a graphics card. An integrated GPU (such as Intel Iris or Apple M-series GPU) is sufficient, although a discrete GPU will provide better performance for shader-based assignments (but not required, our starter code support most of the available GPU series). Several programming exercises will be implemented using real-time GPU shading (GLSL or WebGL), so the machine must support basic GPU compute and graphics APIs.
The course will also include assignments training (very lightweight) neural networks. For these, students will need access to Google Colab, where free GPU compute is available for short sessions (approximately 1 hour per day). Students should ensure they have a Google account and are able to run Python notebooks on Colab.
Basic familiarity with Python and Git is recommended for managing code submissions and collaborating on assignments.
Academic Integrity
All Georgia Tech students are expected to uphold the Georgia Tech Academic Honor Code. This course may impose additional academic integrity stipulations; consult the official course documentation for more information.