Nvidia Corporation

Nividia Corporation

Timo Oskari Aila

Samuli Matias Laine

Tero Tapani Karras

Jaakko Tapani Lehtinen

Miika Samuli Aittala

John Matthew Burgess

William P Newhall, Jr

David Patrick Luebke

Ignacio Llamas

Ronald Charles Babich, Jr

Janne Johannes Hellsten

John Erik Lindholm

Greg Muthler

Ting-Chun Wang

Arun Mohanray Mallya

Ming-Yu Liu

Gregory Muthler

Robert Ohannessian

Craig E Kolb

Ian Kwong

Peter Schuyler Shirley

Michael Garland

Peter Michael Nelson

Carl Jacob Munkberg

Alexander Georg Keller

Jon Niklas Theodor Hasselgren

Erik Andreas Härkönen

Henry Packard Moreton

Jan Kautz

Eric Brian Lum

Yury Y Uralsky

Marco Salvi

Aaron Eliot Lefohn

Anton S Kaplanyan

Morgan McGuire

Pavlo Molchanov

Jacopo Pantaleoni

Jonathan Cohen

Walter Robert Steiner

Nikolaus Binder

Carsten Alexander Waechter

Stephen Walter Tyree

Eric Enderton

David Tarjan

Justin Cobb

Marc Droske

Barry Nolan Rodgers

Daniel Seibert

Mark J Harris

Stefan Radig

Peter Nelson

Shubhabrata Sengupta

Yeongho Seol

Leonhard Gruenschloss

Julia Floetotto

Chakravarty Reddy Alla Chaitanya

Apparatuses, systems, and techniques to perform compression of video data using neural networks to facilitate video streaming, such as video conferencing. In at least one embodiment, a sender transmits to a receiver a key frame from video data and one or more keypoints identified by a neural network from said video data, and a receiver reconstructs video data using said key frame and one or more received keypoints.

Video compression using neural networks

Techniques are disclosed herein for generating a content item. The techniques include receiving a content item and metadata indicating a level of corruption associated with the content item; and for each iteration included in a plurality of iterations: performing one or more operations to add corruption to a first version of the content item to generate a second version of the content item, and performing one or more operations to reduce corruption in the second version of the content item to generate a third version of the content item, wherein a level of corruption associated with the third version of the content item is less than a level of corruption associated with the first version of the content item.

Techniques for content synthesis using denoising diffusion models

Embodiments of the present disclosure relate to a technique for training neural networks, such as a generative adversarial neural network (GAN), using a limited amount of data. Training GANs using too little example data typically leads to discriminator overfitting, causing training to diverge and produce poor results. An adaptive discriminator augmentation mechanism is used that significantly stabilizes training with limited data providing the ability to train high-quality GANs. An augmentation operator is applied to the distribution of inputs to a discriminator used to train a generator, representing a transformation that is invertible to ensure there is no leakage of the augmentations into the images generated by the generator. Reducing the amount of training data that is needed to achieve convergence has the potential to considerably help many applications and may the increase use of generative models in fields such as medicine.

Training neural networks with limited data using invertible augmentation operators

A method and system are disclosed for training a model that implements a machine-learning algorithm. The technique utilizes latent descriptor vectors to change a multiple-valued output problem into a single-valued output problem and includes the steps of receiving a set of training data, processing, by a model, the set of training data to generate a set of output vectors, and adjusting a set of model parameters and component values for at least one latent descriptor vector in the plurality of latent descriptor vectors based on the set of output vectors. The set of training data includes a plurality of input vectors and a plurality of desired output vectors, and each input vector in the plurality of input vectors is associated with a particular latent descriptor vector in a plurality of latent descriptor vectors. Each latent descriptor vector comprises a plurality of scalar values that are initialized prior to training the model.

Machine learning technique for automatic modeling of multiple-valued outputs

Methods and systems are described in some examples for changing the traversal of an acceleration data structure in a highly dynamic query-specific manner, with each query specifying test parameters, a test opcode and a mapping of test results to actions. In an example ray tracing implementation, traversal of a bounding volume hierarchy by a ray is performed with the default behavior of the traversal being changed in accordance with results of a test performed using the test opcode and test parameters specified in the ray data structure and another test parameter specified in a node of the bounding volume hierarchy. In an example implementation a traversal coprocessor is configured to perform the traversal of the bounding volume hierarchy.

Query-specific behavioral modification of tree traversal

Systems and methods are disclosed for fused processing of a continuous mathematical operator. Fused processing of continuous mathematical operations, such as pointwise non-linear functions without storing intermediate results to memory improves performance when the memory bus bandwidth is limited. In an embodiment, a continuous mathematical operation including at least two of convolution, upsampling, pointwise non-linear function, and downsampling is executed to process input data and generate alias-free output data. In an embodiment, the input data is spatially tiled for processing in parallel such that the intermediate results generated during processing of the input data for each tile may be stored in a shared memory within the processor. Storing the intermediate data in the shared memory improves performance compared with storing the intermediate data to the external memory and loading the intermediate data from the external memory.

Fused processing of a continuous mathematical operator

Systems and methods are disclosed that improve output quality of any neural network, particularly an image generative neural network. In the real world, details of different scale tend to transform hierarchically. For example, moving a person's head causes the nose to move, which in turn moves the skin pores on the nose. Conventional generative neural networks do not synthesize images in a natural hierarchical manner: the coarse features seem to mainly control the presence of finer features, but not the precise positions of the finer features. Instead, much of the fine detail appears to be fixed to pixel coordinates which is a manifestation of aliasing. Aliasing breaks the illusion of a solid and coherent object moving in space. A generative neural network with reduced aliasing provides an architecture that exhibits a more natural transformation hierarchy, where the exact sub-pixel position of each feature is inherited from underlying coarse features.

Generative neural networks with reduced aliasing

In a ray tracer, a cache for streaming workloads groups ray requests for coherent successive bounding volume hierarchy traversal operations by sending common data down an attached data path to all ray requests in the group at the same time or about the same time. Grouping the requests provides good performance with a smaller number of cache lines.

Method for efficient grouping of cache requests for datapath scheduling

A hardware-based traversal coprocessor provides acceleration of tree traversal operations searching for intersections between primitives represented in a tree data structure and a ray. The primitives may include triangles used in generating a virtual scene. The hardware-based traversal coprocessor is configured to properly handle numerically challenging computations at or near edges and/or vertices of primitives and/or ensure that a single intersection is reported when a ray intersects a surface formed by primitives at or near edges and/or vertices of the primitives.

Watertight ray triangle intersection

A style-based generative network architecture enables scale-specific control of synthesized output data, such as images. During training, the style-based generative neural network (generator neural network) includes a mapping network and a synthesis network. During prediction, the mapping network may be omitted, replicated, or evaluated several times. The synthesis network may be used to generate highly varied, high-quality output data with a wide variety of attributes. For example, when used to generate images of people's faces, the attributes that may vary are age, ethnicity, camera viewpoint, pose, face shape, eyeglasses, colors (eyes, hair, etc.), hair style, lighting, background, etc. Depending on the task, generated output data may include images, audio, video, three-dimensional (3D) objects, text, etc.

Style-based architecture for generative neural networks

A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Generative adversarial neural network assisted video reconstruction

A generative adversarial neural network (GAN) learns a particular task by being shown many examples. In one scenario, a GAN may be trained to generate new images including specific objects, such as human faces, bicycles, etc. Rather than training a complex GAN having a predetermined topology of features and interconnections between the features to learn the task, the topology of the GAN is modified as the GAN is trained for the task. The topology of the GAN may be simple in the beginning and become more complex as the GAN learns during the training, eventually evolving to match the predetermined topology of the complex GAN. In the beginning the GAN learns large-scale details for the task (bicycles have two wheels) and later, as the GAN becomes more complex, learns smaller details (the wheels have spokes).

Progressive modification of generative adversarial neural networks

A three-dimensional (3D) model of an object is recovered from two-dimensional (2D) images of the object. Each image in the set of 2D images includes the object captured from a different camera position and deformations of a base mesh that defines the 3D model may be computed corresponding to each image. The 3D model may also include a texture map that represents the lighting and material properties of the 3D model. Recovery of the 3D model relies on analytic antialiasing to provide a link between pixel colors in the 2D images and geometry of the 3D model. A modular differentiable renderer design yields high performance by leveraging existing, highly optimized hardware graphics pipelines to reconstruct the 3D model. The differential renderer renders images of the 3D model and differences between the rendered images and reference images are propagated backwards through the rendering pipeline to iteratively adjust the 3D model.

Three-dimensional model recovery from two-dimensional images

Growing community of inventors

Tuusula, Finland

Timo Oskari Aila