Corpus ID: 222133967

Photon-Driven Neural Path Guiding

@article{Zhu2020PhotonDrivenNP,
  title={Photon-Driven Neural Path Guiding},
  author={Shilin Zhu and Zexiang Xu and Tiancheng Sun and Alexandr Kuznetsov and Mark Meyer and Henrik Wann Jensen and Hao Su and Ravi Ramamoorthi},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.01775}
}
Although Monte Carlo path tracing is a simple and effective algorithm to synthesize photo-realistic images, it is often very slow to converge to noise-free results when involving complex global illumination. One of the most successful variance-reduction techniques is path guiding, which can learn better distributions for importance sampling to reduce pixel noise. However, previous methods require a large number of path samples to achieve reliable path guiding. We present a novel neural path… Expand

Figures from this paper

Hierarchical neural reconstruction for path guiding using hybrid path and photon samples
Path guiding is a promising technique to reduce the variance of path tracing. Although existing online path guiding algorithms can eventually learn good sampling distributions given a large amount ofExpand

References

SHOWING 1-10 OF 52 REFERENCES
Offline Deep Importance Sampling for Monte Carlo Path Tracing
TLDR
This paper proposes an offline, scene‐independent deep‐learning approach that can importance sample first‐bounce light paths for general scenes without the need of the costly online training, and can start guiding path sampling with as little as 1 sample per pixel. Expand
Adaptive Incident Radiance Field Sampling and Reconstruction Using Deep Reinforcement Learning
TLDR
This study proposes a method for training quality and reconstruction networks with a massive offline dataset for the adaptive sampling and reconstruction of first-bounce incident radiance fields and demonstrates improvements for unbiased path guiding and competitive performance in biased applications, including filtering and irradiance caching. Expand
Deep Kernel Density Estimation for Photon Mapping
TLDR
This paper develops the first deep learning‐based method for particle‐based rendering that largely reduces the required number of photons, significantly advancing the computational efficiency in photon mapping. Expand
Metropolis light transport
TLDR
A new Monte Carlo method for solving the light transport problem, inspired by the Metropolis sampling method in computational physics, is presented, which is unbiased, handles general geometric and scattering models, uses little storage, and can be orders of magnitude more efficient than previous unbiased approaches. Expand
Practical Path Guiding for Efficient Light‐Transport Simulation
TLDR
This work proposes an adaptive spatio‐directional hybrid data structure, referred to as SD‐tree, for storing and sampling incident radiance, and presents a principled way to automatically budget training and rendering computations to minimize the variance of the final image. Expand
Variance-aware path guiding
TLDR
A generic procedure to derive theoretically optimal target densities for local path guiding is presented and applied in two state-of-the-art rendering applications: a path guiding solution for unidirectional path tracing and a guiding method for light source selection for the many lights problem. Expand
Interactive reconstruction of Monte Carlo image sequences using a recurrent denoising autoencoder
TLDR
This work proposes a variant of deep convolutional networks better suited to the class of noise present in Monte Carlo rendering, which allows for much larger pixel neighborhoods to be taken into account, while also improving execution speed by an order of magnitude. Expand
Stochastic progressive photon mapping
TLDR
Stochastic progressive photon mapping is introduced, which makes it possible to compute the correct average radiance value for a region by using shared photon statistics within the region rather than isolated photon statistics at a point. Expand
On-line learning of parametric mixture models for light transport simulation
TLDR
This work proposes to represent the distributions for sampling scattering directions and light emission by a parametric mixture model trained in an on-line (i.e. progressive) manner from a potentially infinite stream of particles that enables recovering good sampling distributions in scenes with complex lighting. Expand
Product Importance Sampling for Light Transport Path Guiding
TLDR
This work proposes to represent the illumination and the reflectance factors by the Gaussian mixture model (GMM), which is fit by using a combination of weighted expectation maximization and non‐linear optimization methods and shows that it can lead to significantly better convergence in scenes with complex illumination and reflectance. Expand
...
1
2
3
4
5
...