Machine Learning and Rendering
(hovering mouse over a presenter's name shows his bio sketch)
Left: Simple path tracing (left) combined with simple reinforcement learning (middle) outperforms even the much more complicated Metropolis light transport algorithm (right) at the same computational budget. Right: Path guiding based on online learning of parametric mixture models dramatically increases the efficiency of light transport simulation both in simple and complex scenes.
Machine learning techniques just recently enabled dramatic improvements in both realtime and offline rendering. In this course, we introduce the basic principles of machine learning and review their relations to rendering. Besides fundamental facts like the mathematical identity of reinforcement learning and the rendering equation, we cover efficient and surprisingly elegant solutions to light transport simulation, participating media, and noise removal... Extended abstract
AcknowledgementsThe work was supported by the Charles University grant SVV-2017-260452 and by the Czech Science Foundation grant 16-18964S.