公开(公告)号：US10832095B2

公开(公告)日：2020-11-10

申请号：US16164651

申请日：2018-10-18

Applicant: DASSAULT SYSTEMES

Inventor： Jeannine Panciatici ,  Kingdom Torsu ,  Mariem Soltani

IPC: G06K9/62 ,  G06T17/00 ,  G06N20/00 ,  G06K9/00 ,  G06F30/00

Abstract: The disclosure notably relates to a computer-implemented method for forming a dataset configured for learning a function. The function is configured to classify 2D images according to predetermined types of 3D arrangement with respect to objects visible in the 2D images. The method comprising for each respective type of 3D arrangement, constructing 3D scenes each comprising 3D modeled objects arranged according to the respective type of 3D arrangement, generating 2D images each representing a respective 2D perspective of a respective constructed 3D scene where visible 3D modeled objects are among the 3D modeled objects of the respective constructed 3D scene which are arranged according to the respective type of 3D arrangement, and adding to the dataset training patterns each including a respective generated 2D image and information indicative of the respective type of 3D arrangement. Such a method improves 2D image classification.

公开(公告)号：US10831952B2

公开(公告)日：2020-11-10

申请号：US16299460

申请日：2019-03-12

Applicant: Dassault Systemes Simulia Corp.

Inventor： Chenghai Sun ,  Franck Léon Pérot ,  Raoyang Zhang ,  Hudong Chen ,  David M. Freed ,  Ilya Staroselsky ,  Adrien Mann

IPC: G06F17/50 ,  G06F30/23 ,  G06F119/10

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing data representing the effect of tortuosity on the acoustic behavior of a fluid in a porous medium. One of the methods includes generating by a first data processing program of the data processing apparatus, a model of acoustic behavior of a fluid in a porous medium including an effect of tortuosity, with the model comprising a time variable indicative of a sound speed of the fluid. The method includes rescaling the time variable of the model based on the sound speed in a fluid in the porous medium. The method also includes simulating the acoustic behavior including the effect of tortuosity of the porous medium based on the rescaling of the time-related variables within the model.

公开(公告)号：US20200285907A1

公开(公告)日：2020-09-10

申请号：US16879507

申请日：2020-05-20

Applicant: Dassault Systemes

Inventor： Eloi MEHR ,  Andre LIEUTIER

IPC: G06K9/62 ,  G06K9/66 ,  G06K9/00 ,  G06N3/08 ,  G06N3/04 ,  G06T7/30 ,  G06T3/00 ,  G06F30/00

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.

公开(公告)号：US20200250362A1

公开(公告)日：2020-08-06

申请号：US16699829

申请日：2019-12-02

Applicant: Dassault Systemes Simulia Corp.

Inventor： Pradeep Gopalakrishnan ,  Hudong Chen ,  Raoyang Zhang ,  Avinash Jammalamadaka ,  Yanbing Li

IPC: G06F30/23

Abstract: Techniques for simulating fluid flow using a lattice Boltzmann (LB) approach for solving scalar transport equations and solving for total energy are described. In addition to the lattice Boltzmann functions for fluid flow the techniques include modifying a set of state vectors of the particles by adding specific total energy to states of particles that will be advected and subtracting the specific total energy from states of particles that will not be advected over a time interval and performing advection of the particles according to the modified set of states.

公开(公告)号：US20200218838A1

公开(公告)日：2020-07-09

申请号：US16824372

申请日：2020-03-19

Applicant: DASSAULT SYSTEMES

Inventor： Malika BOULKENAFED ,  Philippe Robert Felix BELMANS

IPC: G06F30/20 ,  G06F30/13 ,  G06T17/00 ,  G06T15/00 ,  A63F13/00 ,  G06T19/20 ,  G06T7/11

Abstract: Described is a computer-implemented method for partitioning a 3D scene into a plurality of zones, each zone representing an area or a volume of the 3D scene and being processed by a computing resource. The method comprises obtaining a 3D scene comprising one or more objects, each object generating a computing resource cost, computing a first map that represents a density of computing costs of the provided 3D scene, defining a second map that represents constraints on the shapes of zones that will be obtained as a result of a partitioning of the 3D scene, discretizing the obtained 3D scene into cells by computing a space quantization of the 3D scene free of dynamic objects, computing, for each cell, a computing cost from the first map of the 3D scene, aggregating the cells into one or more zones in accordance with the second map.

公开(公告)号：US20200211276A1

公开(公告)日：2020-07-02

申请号：US16727124

申请日：2019-12-26

Applicant: DASSAULT SYSTEMES

Inventor： Eloi Mehr ,  Fernando Manuel Sanchez Bermudez

IPC: G06T17/00 ,  G06N3/08 ,  G06N3/04

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.

公开(公告)号：US20200210814A1

公开(公告)日：2020-07-02

申请号：US16727092

申请日：2019-12-26

Applicant: DASSAULT SYSTEMES

Inventor： Eloi Mehr

IPC: G06N3/04 ,  G06N20/00 ,  G06N5/04

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 and further includes providing a set of neural networks. Each neural network has respective weights. Each neural network is configured for inference of 3D modeled objects. The method further includes modifying respective weights of the neural networks by minimizing a loss. For each 3D modeled object, the loss selects a term among a plurality of terms. Each term penalizes a disparity between the 3D modeled object and a respective 3D modeled object inferred by a respective neural network of the set. The selected term is a term among the plurality of terms for which the disparity is the least penalized. This constitutes an improved method of machine-learning with a dataset including 3D modeled objects which each represent a respective mechanical part.

公开(公告)号：US20200210632A1

公开(公告)日：2020-07-02

申请号：US16730876

申请日：2019-12-30

Applicant: DASSAULT SYSTEMES

Inventor： Yani SADOUDI ,  Frederic Letzelter

IPC: G06F30/12

Abstract: The disclosure notably relates to a computer-implemented method for designing a three-dimensional (3D) model. The method includes obtaining a first 3D model, the first 3D model being defined by: (i) one delegated data object comprising input parameters specific to a type of the delegated data object and (ii) an output topology, and being associated with a sequence of geometric design operations. The method also includes performing, by a user, a first geometric design operation on the first 3D model, thereby obtaining a second 3D model, determining whether the output topology of the second 3D model can be retrieved from the output topology of the first 3D model, replacing the first delegated data object by a second delegated data object if the output topology of the second 3D model cannot be retrieved from the output topology of the first 3D model or keeping the first delegated data object and storing the first geometric design operation with the sequence of geometric design operations associated the first 3D model.

公开(公告)号：US20200210316A1

公开(公告)日：2020-07-02

申请号：US16730388

申请日：2019-12-30

Applicant: Dassault Systemes Americas Corp.

Inventor： Ryan Cuprak

IPC: G06F11/36 ,  G06F16/901 ,  G06F16/904

Abstract: A computer-implemented method is disclosed that includes receiving content associated with a heap dump of a computer application, generating a plurality of files based on the heap dump content, and loading the files into the graph database. The files so generated are compatible with the graph database. In some implementations, additional analysis and route finding (e.g., finding the relationship between two nodes) may be performed on the resulting object graph.

公开(公告)号：US10671776B2

公开(公告)日：2020-06-02

申请号：US15363958

申请日：2016-11-29

Applicant: Dassault Systemes Simulia Corp.

Inventor： Hudong Chen ,  Rupesh Kotapati ,  Raoyang Zhang ,  Richard Shock ,  Ilya Staroselsky ,  Yanbing Li

IPC: G06F30/23 ,  G06F30/20 ,  G06F17/11 ,  G06F111/10

Abstract: A computer-implemented method for simulating fluid flow using a lattice Boltzmann (LB) approach that includes assigning values for the wall shear stress on a per-facet (e.g., per-surfel) basis based on whether the fluid flow is laminar or turbulent is described herein.