中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

25 records (took 0.021 seconds)

    SYSTEMS AND METHODS FOR TRAINING NEURAL NETWORKS WITH SPARSE DATA
    11.
    发明申请
    SYSTEMS AND METHODS FOR TRAINING NEURAL NETWORKS WITH SPARSE DATA 有权

    公开(公告)号：US20220405582A1

    公开(公告)日：2022-12-22

    申请号：US17665370

    申请日：2022-02-04

    Applicant: NVIDIA Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Niklas Theodor Hasselgren ,  Jaakko T. Lehtinen ,  Timo Oskari Aila

    IPC: G06N3/08 ,  G06N5/04

    Abstract: A method, computer readable medium, and system are disclosed for training a neural network model. The method includes the step of selecting an input vector from a set of training data that includes input vectors and sparse target vectors, where each sparse target vector includes target data corresponding to a subset of samples within an output vector of the neural network model. The method also includes the steps of processing the input vector by the neural network model to produce output data for the samples within the output vector and adjusting parameter values of the neural network model to reduce differences between the output vector and the sparse target vector for the subset of the samples.

    MOTION BLUR AND DEPTH OF FIELD RECONSTRUCTION THROUGH TEMPORALLY STABLE NEURAL NETWORKS
    12.
    发明申请
    MOTION BLUR AND DEPTH OF FIELD RECONSTRUCTION THROUGH TEMPORALLY STABLE NEURAL NETWORKS 审中-公开

    公开(公告)号：US20200051206A1

    公开(公告)日：2020-02-13

    申请号：US16422601

    申请日：2019-05-24

    Applicant: NVIDIA Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Niklas Theodor Hasselgren ,  Marco Salvi

    IPC: G06T3/00 ,  G06T5/00 ,  G06T1/20

    Abstract: A neural network structure, namely a warped external recurrent neural network, is disclosed for reconstructing images with synthesized effects. The effects can include motion blur, depth of field reconstruction (e.g., simulating lens effects), and/or anti-aliasing (e.g., removing artifacts caused by sampling frequency). The warped external recurrent neural network is not recurrent at each layer inside the neural network. Instead, the external state output by the final layer of the neural network is warped and provided as a portion of the input to the neural network for the next image in a sequence of images. In contrast, in a conventional recurrent neural network, hidden state generated at each layer is provided as a feedback input to the generating layer. The neural network can be implemented, at least in part, on a processor. In an embodiment, the neural network is implemented on at least one parallel processing unit.

    SYSTEMS AND METHODS FOR TRAINING NEURAL NETWORKS WITH SPARSE DATA
    13.
    发明申请
    SYSTEMS AND METHODS FOR TRAINING NEURAL NETWORKS WITH SPARSE DATA 审中-公开

    公开(公告)号：US20180357537A1

    公开(公告)日：2018-12-13

    申请号：US15881632

    申请日：2018-01-26

    Applicant: NVIDIA Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Niklas Theodor Hasselgren ,  Jaakko T. Lehtinen ,  Timo Oskari Aila

    IPC: G06N3/08 ,  G06N5/04

    CPC classification number: G06N3/08 ,  G06N5/04

    Abstract: A method, computer readable medium, and system are disclosed for training a neural network model. The method includes the step of selecting an input vector from a set of training data that includes input vectors and sparse target vectors, where each sparse target vector includes target data corresponding to a subset of samples within an output vector of the neural network model. The method also includes the steps of processing the input vector by the neural network model to produce output data for the samples within the output vector and adjusting parameter values of the neural network model to reduce differences between the output vector and the sparse target vector for the subset of the samples.

    NOISE-FREE DIFFERENTIABLE RAY CASTING
    14.
    发明公开
    NOISE-FREE DIFFERENTIABLE RAY CASTING 审中-公开

    公开(公告)号：US20230316631A1

    公开(公告)日：2023-10-05

    申请号：US17840791

    申请日：2022-06-15

    Applicant: NVIDIA Corporation

    Inventor： Jon Niklas Theodor Hasselgren ,  Carl Jacob Munkberg

    IPC: G06T15/06 ,  G06T15/00 ,  G06T7/13

    CPC classification number: G06T15/06 ,  G06T7/13 ,  G06T15/005

    Abstract: A differentiable ray casting technique may be applied to a model of a three-dimensional (3D) scene (scene includes lighting configuration) or object to optimize one or more parameters of the model. The one or more parameters define geometry (topology and shape), materials, and lighting configuration (e.g., environment map, a high-resolution texture that represents the light coming from all directions in a sphere) for the model. Visibility is computed in 3D space by casting at least two rays from each ray origin (where the two rays define a ray cone). The model is rendered to produce a model image that may be compared with a reference image (or photograph) of a reference 3D scene to compute image space differences. Visibility gradients in 3D space are computed and backpropagated through the computations to reduce differences between the model image and the reference image.

    Appearance-driven automatic three-dimensional modeling
    15.
    发明授权
    Appearance-driven automatic three-dimensional modeling 有权

    公开(公告)号：US11450077B2

    公开(公告)日：2022-09-20

    申请号：US17194477

    申请日：2021-03-08

    Applicant: NVIDIA Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Niklas Theodor Hasselgren

    IPC: G06T19/20 ,  G06T7/55 ,  G06T1/20 ,  G06T15/20 ,  G06T13/20

    Abstract: Appearance driven automatic three-dimensional (3D) modeling enables optimization of a 3D model comprising the shape and appearance of a particular 3D scene or object. Triangle meshes and shading models may be jointly optimized to match the appearance of a reference 3D model based on reference images of the reference 3D model. Compared with the reference 3D model, the optimized 3D model is a lower resolution 3D model that can be rendered in less time. More specifically, the optimized 3D model may include fewer geometric primitives compared with the reference 3D model. In contrast with the conventional inverse rendering or analysis-by-synthesis modeling tools, the shape and appearance representations of the 3D model are automatically generated that, when rendered, match the reference images. Appearance driven automatic 3D modeling has a number of uses, including appearance-preserving simplification of extremely complex assets, conversion between rendering systems, and even conversion between geometric scene representations.

    Systems and methods for training neural networks with sparse data
    16.
    发明授权
    Systems and methods for training neural networks with sparse data 有权

    公开(公告)号：US11244226B2

    公开(公告)日：2022-02-08

    申请号：US15881632

    申请日：2018-01-26

    Applicant: NVIDIA Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Niklas Theodor Hasselgren ,  Jaakko T. Lehtinen ,  Timo Oskari Aila

    IPC: G06N3/08 ,  G06N5/04

    Abstract: A method, computer readable medium, and system are disclosed for training a neural network model. The method includes the step of selecting an input vector from a set of training data that includes input vectors and sparse target vectors, where each sparse target vector includes target data corresponding to a subset of samples within an output vector of the neural network model. The method also includes the steps of processing the input vector by the neural network model to produce output data for the samples within the output vector and adjusting parameter values of the neural network model to reduce differences between the output vector and the sparse target vector for the subset of the samples.

    MOTION BLUR AND DEPTH OF FIELD RECONSTRUCTION THROUGH TEMPORALLY STABLE NEURAL NETWORKS
    18.
    发明申请
    MOTION BLUR AND DEPTH OF FIELD RECONSTRUCTION THROUGH TEMPORALLY STABLE NEURAL NETWORKS 有权

    公开(公告)号：US20210264562A1

    公开(公告)日：2021-08-26

    申请号：US17213941

    申请日：2021-03-26

    Applicant: Nvidia Corporation

    Inventor： Carl Jacob Munkberg ,  Jon Hasselgren ,  Marco Salvi

    IPC: G06T3/00 ,  G06T5/00 ,  G06T1/20

    Abstract: A neural network structure, namely a warped external recurrent neural network, is disclosed for reconstructing images with synthesized effects. The effects can include motion blur, depth of field reconstruction (e.g., simulating lens effects), and/or anti-aliasing (e.g., removing artifacts caused by sampling frequency). The warped external recurrent neural network is not recurrent at each layer inside the neural network. Instead, the external state output by the final layer of the neural network is warped and provided as a portion of the input to the neural network for the next image in a sequence of images. In contrast, in a conventional recurrent neural network, hidden state generated at each layer is provided as a feedback input to the generating layer. The neural network can be implemented, at least in part, on a processor. In an embodiment, the neural network is implemented on at least one parallel processing unit.

Patent Agency Ranking