Nvidia Corporation

Shiqiu Liu

Jacopo Pantaleoni

Andrew J Tao

Jon N Hasselgren

Ignacio Llamas

Christopher Ryan Wyman

Matthieu Le

Jacob Munkberg

Nan Lin

Yaobin Ouyang

Markus Kettunen

Bryan Catanzaro

Ting-Chun Wang

Zhiding Yu

Guilin Liu

Karan Sapra

Fitsum A Reda

Brandon Rowlett

Disclosed are apparatuses, systems, and techniques to render images with global illumination using efficient ray tracing, light source identification, and reservoir resampling that deploys temporal and spatial reservoirs.

Ray tracing using reservoir resampling with spatial shift-mapping

Various approaches are disclosed to temporally and spatially filter noisy image data—generated using one or more ray-tracing effects—in a graphically rendered image. Rather than fully sampling data values using spatial filters, the data values may be sparsely sampled using filter taps within the spatial filters. To account for the sparse sampling, locations of filter taps may be jittered spatially and/or temporally. For filtering efficiency, a size of a spatial filter may be reduced when historical data values are used to temporally filter pixels. Further, data values filtered using a temporal filter may be clamped to avoid ghosting. For further filtering efficiency, a spatial filter may be applied as a separable filter in which the filtering for a filter direction may be performed over multiple iterations using reducing filter widths, decreasing the chance of visual artifacts when the spatial filter does not follow a true Gaussian distribution.

Rendering frames using jittered filter taps

Apparatuses, systems, and techniques for texture synthesis from small input textures in images using convolutional neural networks. In at least one embodiment, one or more convolutional layers are used in conjunction with one or more transposed convolution operations to generate a large textured output image from a small input textured image while preserving global features and texture, according to various novel techniques described herein.

Techniques to use a neural network to expand an image

Apparatuses, systems, and techniques to enhance video are disclosed. In at least one embodiment, one or more neural networks are used to create a higher resolution video using upsampled frames from a lower resolution video.

Video upsampling using one or more neural networks

Disclosed approaches may leverage the actual spatial and reflective properties of a virtual environment—such as the size, shape, and orientation of a bidirectional reflectance distribution function (BRDF) lobe of a light path and its position relative to a reflection surface, a virtual screen, and a virtual camera—to produce, for a pixel, an anisotropic kernel filter having dimensions and weights that accurately reflect the spatial characteristics of the virtual environment as well as the reflective properties of the surface. In order to accomplish this, geometry may be computed that corresponds to a projection of a reflection of the BRDF lobe below the surface along a view vector to the pixel. Using this approach, the dimensions of the anisotropic filter kernel may correspond to the BRDF lobe to accurately reflect the spatial characteristics of the virtual environment as well as the reflective properties of the surface.

Reflection denoising in ray-tracing applications

Various approaches are disclosed to temporally and spatially filter noisy image data&#x2014;generated using one or more ray-tracing effects&#x2014;in a graphically rendered image. Rather than fully sampling data values using spatial filters, the data values may be sparsely sampled using filter taps within the spatial filters. To account for the sparse sampling, locations of filter taps may be jittered spatially and/or temporally. For filtering efficiency, a size of a spatial filter may be reduced when historical data values are used to temporally filter pixels. Further, data values filtered using a temporal filter may be clamped to avoid ghosting. For further filtering efficiency, a spatial filter may be applied as a separable filter in which the filtering for a filter direction may be performed over multiple iterations using reducing filter widths, decreasing the chance of visual artifacts when the spatial filter does not follow a true Gaussian distribution.

Filtering render data using multiple iterations for a filter direction

This disclosure presents a method and computer program product to denoise a ray traced scene. An apparatus for processing a ray traced scene is also disclosed. In one example, the method includes: (1) generating filtered scene data by filtering modified scene data from original scene data utilizing a spatial filter, and (2) providing a denoised ray traced scene by adjusting the filtered scene data utilizing a temporal filter. The modified and adjusted scene data can be sent to a rendering processor or system to complete rendering to generate a final scene.

Using previously rendered scene frames to reduce pixel noise

Disclosed approaches may leverage the actual spatial and reflective properties of a virtual environment&#x2014;such as the size, shape, and orientation of a bidirectional reflectance distribution function (BRDF) lobe of a light path and its position relative to a reflection surface, a virtual screen, and a virtual camera&#x2014;to produce, for a pixel, an anisotropic kernel filter having dimensions and weights that accurately reflect the spatial characteristics of the virtual environment as well as the reflective properties of the surface. In order to accomplish this, geometry may be computed that corresponds to a projection of a reflection of the BRDF lobe below the surface along a view vector to the pixel. Using this approach, the dimensions of the anisotropic filter kernel may correspond to the BRDF lobe to accurately reflect the spatial characteristics of the virtual environment as well as the reflective properties of the surface.

In various examples, the actual spatial properties of a virtual environment are used to produce, for a pixel, an anisotropic filter kernel for a filter having dimensions and weights that accurately reflect the spatial characteristics of the virtual environment. Geometry of the virtual environment may be computed based at least in part on a projection of a light source onto a surface through an occluder, in order to determine a footprint that reflects a contribution of the light source to lighting conditions of the pixel associated with a point on the surface. The footprint may define a size, orientation, and/or shape of the anisotropic filter kernel and corresponding filter weights. The anisotropic filter kernel may be applied to the pixel to produce a graphically-rendered image of the virtual environment.

Shadow denoising in ray-tracing applications

Temporal-spatial denoising in ray-tracing applications

This disclosure presents a method to denoise a ray traced scene where the ray tracing uses a minimal number of rays. The method can use temporal reprojections to compute a weighted average to the scene data. A spatial filter can be run on the scene data, using the temporal reprojection count to reduce the size of the utilized spatial filter radius. In some aspects, additional temporal filters can be applied to the scene data. In some aspects, global illumination temporal reprojection history counts can be used to modify the spatial filter radius. In some aspects, caustic photon tracing can be conducted to compute a logarithmic cost, which can then be utilized to reduce the denoising radius used by the spatial filter. The modified and adjusted scene data can be sent to a rendering process to complete the rendering to generate a final scene.

Growing community of inventors

Santa Clara, CA, United States of America

Shiqiu Liu