Stanford CS348B, Spring 2022

High-quality rendering is ubiquitous today, with applications ranging from entertainment to product design and architecture. The goal of this course is to provide a deep understanding of the fundamental mathematical and physical principles that are the basis of modern physically based rendering while also introducing the design principles and engineering trade-offs involved in designing and implementing high-performance rendering systems.

Basic Info
Tues/Thurs 3:15-4:45pm
Location: 380-380F
Instructors: Kayvon Fatahalian, Doug James, and Matt Pharr
See the course info page for more info on policies and logistics.
Spring 2022 Schedule
Mar 29
How advanced image synthesis is used in the real world, review of ray tracing basics
Mar 31
Representing rays, ray-object intersection methods, acceleration structures, pbrt basics
Apr 05
how to build acceleration structures efficiently (two-level, refitting, incremental builds), advanced primitive types, tessellation, numerical precision issues
Apr 07
Definition of radiometric quantities, the light field, integrating total energy falling on surfaces
Apr 12
Sampling from distributions and shapes, numerical estimation of illumination
Apr 14
Basics of how lenses and sensors work, motion blur and depth of field
Apr 19
Primal and Fourier space representations of signals, convolution theorem, sampling theorem, aliasing and anti-aliasing
Apr 21
BRDFs, the reflection equation, basic reflection models
Apr 26
Advanced surface models
Apr 28
Monte Carlo estimation of the reflection equation, sampling lights and BRDFs
May 03
Monte Carlo variance reduction techniques, stratified sampling, importance sampling
May 05
The rendering equation, path tracing, Russian roulette, path guiding
May 10
Light tracing, bidirectional path tracing, photon mapping
May 12
Scattering and phase functions, the volume rendering equation, null scattering
May 17
Final Project Proposals
Slide presentations and discussion
May 19
Anisotropic surfaces, subsurface scattering
May 24
Algorithms for reducing variance at low sampling count (ReSTIR), conventional and DNN-based denoising
May 26
Discrepancy and Quasi-Monte Carlo, low-discrepancy constructions, spectral analysis of sampling
May 31
Three parts: (1) guided sampling [Kayvon & Vishnu], (2) warped rendering [Doug], and (3) course wrap-up [Matt] (on-going research and how to get involved).
Apr 12 HW 1: PBRT Lighting Design
Apr 21 HW 2: A Ray Marching-Based Distance Estimator
May 3 HW 3: Light Field Cameras