Denoising Monte Carlo renderings using machine learning with importance sampling
US10789686B2 · kind B2 · utility
Assignees
Inventors
Key dates
| Filing date | Jan 6, 2020 |
| Grant date | Sep 29, 2020 |
| Priority date | — |
| Expiry date | Jan 6, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30201
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Supervised machine learning using neural networks is applied to denoising images rendered by MC path tracing. Specialization of neural networks may be achieved by using a modular design that allows reusing trained components in different networks and facilitates easy debugging and incremental building of complex structures. Specialization may also be achieved by using progressive neural networks. In some embodiments, training of a neural-network based denoiser may use importance sampling, where more challenging patches or patches including areas of particular interests within a training dataset are selected with higher probabilities than others. In some other embodiments, generative adversarial networks (GANs) may be used for training a machine-learning based denoiser as an alternative to using pre-defined loss functions.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.