Neural Graphics
Course ID 15474
Description This course explores the rapidly evolving intersection of machine learning and computer graphics, offering a modern perspective on topics traditionally covered in 15- 362. Students will learn how neural methods are transforming the way we represent, render, and animate visual content. Through a combination of lectures, seminar-style discussions, and research-based projects, the course examines how machine learning techniques can accelerate and enhance traditional graphics pipelines. Topics include Neural Texture Compression, Computer Shaders, Neural Shading, Neural Denoising, Gaussian Splatting, Neural Radiance Fields, Inverse Rendering, AI Players, Animation Cycles, Neural Frame Interpolation, & Stylized Animation.
Required Background Knowledge
Computer graphics/vision knowledge equivalent to 15-362 and/or 16-385. Machine
Learning knowledge equivalent to 15-301 or higher strongly preferred. Students with a
strong machine learning and/or graphics background who have not met the course pre-
requisites may contact the instructor for permission to enroll.