Disney Enterprises, Inc.

Pixar

Eth Zurich (eidgenossische Technische Hochschule Zurich)

Eth Zurich

Disnev Enterprises, Inc.

Brian McWilliams

Thijs Vogels

Fabrice Rousselle

Jan Novak

Mark Joseph Meyer

Alex Harvill

Alexander Sorkine Hornung

Christopher Richard Schroers

Jan Novák

Markus H Gross

Federico Perazzi

Abdelaziz Djelouah

Fabrice Pierre Armand Rousselle

Simone Meyer

Yifan Wang

A modular architecture is provided for denoising Monte Carlo renderings using neural networks. The temporal approach extracts and combines feature representations from neighboring frames rather than building a temporal context using recurrent connections. A multiscale architecture includes separate single-frame or temporal denoising modules for individual scales, and one or more scale compositor neural networks configured to adaptively blend individual scales. An error-predicting module is configured to produce adaptive sampling maps for a renderer to achieve more uniform residual noise distribution. An asymmetric loss function may be used for training the neural networks, which can provide control over the variance-bias trade-off during denoising.

Temporal techniques of denoising Monte Carlo renderings using neural networks

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.

Denoising Monte Carlo renderings using progressive neural networks

According to one implementation, a system includes a computing platform having a hardware processor and a system memory storing a software code including multiple artificial neural networks (ANNs). The hardware processor executes the software code to partition a multi-dimensional input vector into a first vector data and a second vector data, and to transform the second vector data using a first piecewise-polynomial transformation parameterized by one of the ANNs, based on the first vector data, to produce a transformed second vector data. The hardware processor further executes the software code to transform the first vector data using a second piecewise-polynomial transformation parameterized by another of the ANNs, based on the transformed second vector data, to produce a transformed first vector data, and to determine a multi-dimensional output vector based on an output from the plurality of ANNs.

Piecewise-polynomial coupling layers for warp-predicting neural networks

Supervised machine learning using convolutional neural network (CNN) is applied to denoising images rendered by MC path tracing. The input image data may include pixel color and its variance, as well as a set of auxiliary buffers that encode scene information (e.g., surface normal, albedo, depth, and their corresponding variances). In some embodiments, a CNN directly predicts the final denoised pixel value as a highly non-linear combination of the input features. In some other embodiments, a kernel-prediction neural network uses a CNN to estimate the local weighting kernels, which are used to compute each denoised pixel from its neighbors. In some embodiments, the input image can be decomposed into diffuse and specular components. The diffuse and specular components are then independently preprocessed, filtered, and postprocessed, before recombining them to obtain a final denoised image.

Kernel-predicting convolutional neural networks for denoising

Denoising Monte Carlo renderings using machine learning with importance sampling

Adaptive sampling in Monte Carlo renderings using error-predicting neural networks

Denoising Monte Carlo renderings using neural networks with asymmetric loss

Multi-scale architecture of denoising monte carlo renderings using neural networks

According to one implementation, a video processing system includes a computing platform having a hardware processor and a system memory storing a software code including an artificial neural network (ANN). The hardware processor is configured to execute the software code to receive a first video sequence having a first display resolution, and to produce a second video sequence based on the first video sequence using the ANN. The second video sequence has a second display resolution higher than the first display resolution. The ANN is configured to provide sequential frames of the second video sequence that are temporally stable and consistent in color to reduce visual flicker and color shifting in the second video sequence.

Video super-resolution using an artificial neural network

Denoising monte carlo renderings using progressive neural networks

Denoising Monte Carlo renderings using generative adversarial neural networks

According to one implementation, a video processing system includes a computing platform having a hardware processor and a system memory storing a frame interpolation software code, the frame interpolation software code including a convolutional neural network (CNN) trained using a loss function having an image loss term summed with a phase loss term. The hardware processor executes the frame interpolation software code to receive first and second consecutive video frames including respective first and second images, and to decompose the first and second images to produce respective first and second image decompositions. The hardware processor further executes the frame interpolation software code to use the CNN to determine an intermediate image decomposition corresponding to an interpolated video frame for insertion between the first and second video frames based on the first and second image decompositions, and to synthesize the interpolated video frame based on the intermediate image decomposition.

Video frame interpolation using a convolutional neural network

Growing community of inventors

Zurich, Switzerland

Brian McWilliams