中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

23 records (took 0.013 seconds)

    Determining an architectural layout
    11.
    发明授权
    Determining an architectural layout 有权

    公开(公告)号：US10832079B2

    公开(公告)日：2020-11-10

    申请号：US15975407

    申请日：2018-05-09

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Adil Baaj

    IPC: G06K9/46 ,  G06K9/66 ,  G06T17/00 ,  G06F30/13 ,  G06T7/143 ,  G06F17/18

    Abstract: A computer-implemented method for determining an architectural layout. The method comprises providing a cycle of points that represents a planar cross section of a cycle of walls, and, assigned to each respective point, a respective first datum that represents a direction normal to the cycle of points at the respective point. The method also comprises minimizing a Markov Random Field energy thereby assigning, to each respective point, a respective one of the set of second data. The method also comprises identifying maximal sets of consecutive points to which a same second datum is assigned, and a cycle of vertices bounding a cycle of segments which represents the architectural layout. Such a method constitutes an improved solution for determining an architectural layout.

    FORMING A DATASET FOR INFERENCE OF EDITABLE FEATURE TREES
    12.
    发明申请
    FORMING A DATASET FOR INFERENCE OF EDITABLE FEATURE TREES 审中-公开

    公开(公告)号：US20200250894A1

    公开(公告)日：2020-08-06

    申请号：US16727207

    申请日：2019-12-26

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Fernando Manuel Sanchez Bermudez

    IPC: G06T19/20 ,  G06N3/08 ,  G06F30/12 ,  G06F16/901

    Abstract: The disclosure notably relates to a computer-implemented method for forming a dataset configured for learning a neural network. The neural network is configured for inference, from a discrete geometrical representation of a 3D shape, of an editable feature tree representing the 3D shape. The editable feature tree comprises a tree arrangement of geometrical operations applied to leaf geometrical shapes. The method includes obtaining respective data pieces, and inserting a part of the data pieces in the dataset each as a respective training sample. The respective 3D shape of each of one or more first data pieces inserted in the dataset is identical to the respective 3D shape of respective one or more second data pieces not inserted in the dataset. The method forms an improved solution for digitization.

    Augmenting a video flux of a real scene
    14.
    发明授权
    Augmenting a video flux of a real scene 有权

    公开(公告)号：US11631221B2

    公开(公告)日：2023-04-18

    申请号：US17138259

    申请日：2020-12-30

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Vincent Guitteny

    IPC: G06T19/00 ,  G06T7/73 ,  G06K9/62

    Abstract: A computer-implemented method of augmented reality includes capturing the video flux with a video camera, extracting, from the video flux, one or more 2D images each representing the real object, and obtaining a 3D model representing the real object. The method also includes determining a pose of the 3D model relative to the video flux, among candidate poses. The determining rewards a mutual information, for at least one 2D image and for each given candidate pose, which represents a mutual dependence between a virtual 2D rendering and the at least one 2D image. The method also includes augmenting the video flux based on the pose. This forms an improved solution of augmented reality for augmenting a video flux of a real scene including a real object.

    Machine-learning for 3D modeled object inference
    15.
    发明授权
    Machine-learning for 3D modeled object inference 有权

    公开(公告)号：US11443192B2

    公开(公告)日：2022-09-13

    申请号：US16727035

    申请日：2019-12-26

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr

    IPC: G06N3/08 ,  G06T7/55 ,  G06T17/20 ,  G06N20/00

    Abstract: The disclosure notably relates to a computer-implemented method of machine-learning. The method includes obtaining a dataset including 3D modeled objects which each represent a respective mechanical part. The dataset has one or more sub-datasets. Each sub-dataset forms at least a part of the dataset. The method further includes, for each respective sub-dataset, determining a base template and learning a neural network configured for inference of deformations of the base template each into a respective 3D modeled object. The base template is a 3D modeled object which represents a centroid of the 3D modeled objects of the sub-dataset. The learning includes a training based on the sub-dataset. This constitutes an improved method of machine-learning with a dataset including 3D modeled objects which each represent a respective mechanical part.

    Learning a neural network for inference of editable feature trees
    16.
    发明授权
    Learning a neural network for inference of editable feature trees 有权

    公开(公告)号：US11436795B2

    公开(公告)日：2022-09-06

    申请号：US16727124

    申请日：2019-12-26

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Fernando Manuel Sanchez Bermudez

    IPC: G06T17/00 ,  G06N3/04 ,  G06N3/08 ,  G06N3/00 ,  G06T17/10 ,  G06F3/0484 ,  G06N5/04

    Abstract: The disclosure notably relates to a computer-implemented method for learning a neural network configured for inference, from a discrete geometrical representation of a 3D shape, of an editable feature tree representing the 3D shape. The editable feature tree includes a tree arrangement of geometrical operations applied to leaf geometrical shapes. The method includes obtaining a dataset including discrete geometrical representations each of a respective 3D shape, and obtaining a candidate set of leaf geometrical shapes. The method also includes learning the neural network based on the dataset and on the candidate set. The candidate set includes at least one continuous subset of leaf geometrical shapes. The method forms an improved solution for digitization.

    Texturing a 3D modeled object
    17.
    发明授权
    Texturing a 3D modeled object 有权

    公开(公告)号：US10013801B2

    公开(公告)日：2018-07-03

    申请号：US14949686

    申请日：2015-11-23

    Applicant: Dassault Systemes

    Inventor： Eloi Mehr

    IPC: G06T17/20 ,  G06T15/04 ,  G06F17/18 ,  G06F17/50 ,  G06T15/10 ,  G06F17/11 ,  G06T7/50

    CPC classification number: G06T17/20 ,  G06F17/11 ,  G06F17/18 ,  G06F17/50 ,  G06T7/50 ,  G06T15/04 ,  G06T15/10 ,  G06T2207/10028 ,  G06T2207/20076

    Abstract: A computer-implemented method for designing a 3D modeled object representing a real object comprises providing a 3D mesh representing the real object, a texturing image and a mapping between the vertices of the 3D mesh and pixels of the texturing image; then maximizing a probability P(L(V)) of the form: P ⁡ ( L ⁡ ( V ) ) = 1 Z ⁢ exp ⁡ ( - ∑ i = 1 n ⁢ ⁢ φ i ′ ⁡ ( L ⁡ ( v i ) ) - ∑ f ∈ ℱ ⁢ ⁢ ψ f ′ ⁡ ( { L ⁡ ( v i ) } i ∈ f ) ) . Maximizing is performed with a predetermined discrete Markov Random Field optimization scheme viewing the 3D mesh and the pixel shifts associated to the texture coordinates of the vertices of the 3D mesh as a Markov Random Field of energy −log(P(L(V)))−log(Z). The method then comprises texturing the 3D mesh according to the texturing image, to the mapping, and to the result of the maximizing. This provides an improved solution for designing a 3D modeled object a real object.

    Learning an autoencoder
    18.
    发明授权
    Learning an autoencoder 有权

    公开(公告)号：US11468268B2

    公开(公告)日：2022-10-11

    申请号：US16879507

    申请日：2020-05-20

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Andre Lieutier

    IPC: G06K9/62 ,  G06N3/08 ,  G06N3/04 ,  G06T7/30 ,  G06T3/00 ,  G06F30/00 ,  G06V10/75 ,  G06V20/64 ,  G06V30/194 ,  G06F111/04

    Abstract: A computer-implemented method for learning an autoencoder notably is provided. The method includes obtaining a dataset of images. Each image includes a respective object representation. The method also includes learning the autoencoder based on the dataset. The learning includes minimization of a reconstruction loss. The reconstruction loss includes a term that penalizes a distance for each respective image. The penalized distance is between the result of applying the autoencoder to the respective image and the set of results of applying at least part of a group of transformations to the object representation of the respective image. Such a method provides an improved solution to learn an autoencoder.

    Forming a dataset for inference of editable feature trees
    19.
    发明授权
    Forming a dataset for inference of editable feature trees 有权

    公开(公告)号：US11210866B2

    公开(公告)日：2021-12-28

    申请号：US16727207

    申请日：2019-12-26

    Applicant: DASSAULT SYSTEMES

    Inventor： Eloi Mehr ,  Fernando Manuel Sanchez Bermudez

    IPC: G06T19/20 ,  G06F16/901 ,  G06F30/12 ,  G06N3/08 ,  G06T17/10 ,  G06N3/04 ,  G06N5/00 ,  G06T17/00 ,  G06F9/451

    Abstract: The disclosure notably relates to a computer-implemented method for forming a dataset configured for learning a neural network. The neural network is configured for inference, from a discrete geometrical representation of a 3D shape, of an editable feature tree representing the 3D shape. The editable feature tree comprises a tree arrangement of geometrical operations applied to leaf geometrical shapes. The method includes obtaining respective data pieces, and inserting a part of the data pieces in the dataset each as a respective training sample. The respective 3D shape of each of one or more first data pieces inserted in the dataset is identical to the respective 3D shape of respective one or more second data pieces not inserted in the dataset. The method forms an improved solution for digitization.

Patent Agency Ranking