Method generates adjustable smooth noise for time based tasks
Tabula Rasa: Monte Carlo estimation of unit-variance noise with controlled spatio-temporal correlation
GraphicsComputer Vision and Pattern Recognition
Summary
Sometimes computers need to create noise, like a static sound or grainy image, that changes smoothly over time and has certain patterns. The authors came up with a new way to make this noise have both precise strength and smooth changes from one moment to the next. They use a clever math trick called Monte Carlo estimation combined with a technique called sketching to do this efficiently. Their approach makes it easier and faster to generate such noise for tasks that depend on consistent timing.
What this means in practice
- •For graphics developers: Create time-varying visual noise with controlled smoothness for animations and rendering effects.
- •For signal processing teams: Generate noise signals with precise variance and timing patterns for testing and control applications.
Authors
Tobias Ritschel, Yang Zhou, Nick Milef, Mikhail Dereviannykh, Chen Liu, Christophe Hery, Carl Marshall
Abstract
We suggest a method to generate time-varying Gaussian noise with controlled variance and controlled temporal correlation. This noise is used in several downstream tasks for temporal control and temporal coherence. The core technical idea is to phrase this problem as joint Monte-Carlo estimation of both a classic pixel reconstruction and estimation of variance using the concept of "sketching" from the database literature. We demonstrate that our method allows temporal control for downstream tasks with simpler and faster code than previous methods.