中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

14 records (took 0.017 seconds)

    DEPTH MAP GENERATION
    11.
    发明申请
    DEPTH MAP GENERATION 有权
    深度地图生成

    公开(公告)号：US20160163053A1

    公开(公告)日：2016-06-09

    申请号：US15046021

    申请日：2016-02-17

    Applicant: Adobe Systems Incorporated

    Inventor： Huixuan Tang ,  Scott Cohen ,  Stephen Schiller ,  Brian Price

    IPC: G06T7/00 ,  G06K9/62 ,  H04N17/00 ,  G06K9/46

    CPC classification number: H04N13/128 ,  G06K9/4661 ,  G06K9/6215 ,  G06K2009/4666 ,  G06T5/003 ,  G06T5/50 ,  G06T7/571 ,  G06T2207/10016 ,  G06T2207/10028 ,  G06T2207/10148 ,  G06T2207/20048 ,  H04N1/387 ,  H04N5/2226 ,  H04N5/23212 ,  H04N5/23222 ,  H04N5/2329 ,  H04N5/2351 ,  H04N5/2355 ,  H04N5/33 ,  H04N13/271 ,  H04N17/002 ,  H04N2013/0081 ,  H04N2213/003

    Abstract: Depth maps are generated from two or more of images captured with a conventional digital camera from the same viewpoint using different configuration settings, which may be arbitrarily selected for each image. The configuration settings may include aperture and focus settings and/or other configuration settings capable of introducing blur into an image. The depth of a selected image patch is evaluated over a set of discrete depth hypotheses using a depth likelihood function modeled to analyze corresponding images patches convolved with blur kernels using a flat prior in the frequency domain. In this way, the depth likelihood function may be evaluated without first reconstructing an all-in-focus image. Blur kernels used in the depth likelihood function and are identified from a mapping of depths and configuration settings to the blur kernels. This mapping is determined from calibration data for the digital camera used to capture the two or more images.

    Abstract translation: 使用不同的配置设置从相同的观点利用传统的数码相机捕获的两个或更多个图像生成深度贴图，这可以为每个图像任意选择。 配置设置可以包括能够将模糊引入到图像中的孔径和焦点设置和/或其他配置设置。 使用深度似然函数对所选图像块的深度进行评估，所述深度假设用模型来分析在频域中使用平坦先验的与模糊粒子卷积的相应图像片段。 以这种方式，可以在不首先重建全焦点图像的情况下评估深度似然函数。 在深度似然函数中使用的模糊内核，并从深度和配置设置到模糊内核的映射中识别。 该映射由用于捕获两个或更多个图像的数字照相机的校准数据确定。

    Image matting using deep learning
    12.
    发明授权
    Image matting using deep learning 有权

    公开(公告)号：US10255681B2

    公开(公告)日：2019-04-09

    申请号：US15448541

    申请日：2017-03-02

    Applicant: ADOBE SYSTEMS INCORPORATED

    Inventor： Brian Lynn Price ,  Stephen Schiller ,  Scott Cohen ,  Ning Xu

    IPC: G06K9/36 ,  G06K9/46 ,  G06N3/08 ,  G06T7/90 ,  G06T7/194

    Abstract: Methods and systems are provided for generating mattes for input images. A neural network system can be trained where the training includes training a first neural network that generates mattes for input images where the input images are synthetic composite images. Such a neural network system can further be trained where the training includes training a second neural network that generates refined mattes from the mattes produced by the first neural network. Such a trained neural network system can be used to input an image and trimap pair for which the trained system will output a matte. Such a matte can be used to extract an object from the input image. Upon extracting the object, a user can manipulate the object, for example, to composite the object onto a new background.

    Enhancing curves using non-uniformly scaled cubic variation of curvature curves
    13.
    发明授权
    Enhancing curves using non-uniformly scaled cubic variation of curvature curves 有权

    公开(公告)号：US09984480B2

    公开(公告)日：2018-05-29

    申请号：US15076423

    申请日：2016-03-21

    Applicant: Adobe Systems Incorporated

    Inventor： Stephen Schiller ,  Nathan Carr

    IPC: G06T11/20 ,  G06F17/24

    CPC classification number: G06T11/203 ,  G06F17/242

    Abstract: The present disclosure is directed to generating enhanced curves that are aesthetically pleasing. To create enhanced a curve that is aesthetically pleasing, a curve enhancement system uses non-uniformly scaled cubic variation of curvature (CVC) curves. For example, the curve enhancement system non-uniformly scales a curve in a spline. Based on the scaling, the curve enhancement system can generate CVC curves having the desired end point constraints. Then, using the end point constraints, the curve enhancement system can inversely downscale the non-uniform scaled curve while maintaining the end point constraints from the CVC curves to achieve an enhanced curve in the spline.

Patent Agency Ranking