公开(公告)号：US10586337B2

公开(公告)日：2020-03-10

申请号：US15857821

申请日：2017-12-29

Applicant: DASSAULT SYSTEMES

Inventor： Fernando Sanchez Bermudez ,  Eloi Mehr ,  David Bonner ,  Vincent Guitteny ,  Mourad Boufarguine ,  Patrick Johnson

IPC: G06T7/143 ,  G06T7/174 ,  G06T7/00 ,  G06T7/187

Abstract: A computer-implemented method of computer vision in a scene that includes one or more transparent objects and/or one or more reflecting objects comprises obtaining a plurality of images of the scene, each image corresponding to a respective acquisition of a physical signal, the plurality of images including at least two images corresponding to different physical signals; and generating a segmented image of the scene based on the plurality of images. This improves the field of computer vision.

公开(公告)号：US10783707B2

公开(公告)日：2020-09-22

申请号：US16230841

申请日：2018-12-21

Applicant: Dassault Systemes

Inventor： Fernando Manuel Sanchez Bermudez ,  Mourad Boufarguine ,  Guillaume Randon

IPC: G06T17/20 ,  G06T17/30 ,  G06F30/00

Abstract: The disclosure notably relates to a computer-implemented method for 3D reconstruction. The method comprises providing a 3D point cloud representing a real object. The method also comprises fitting the 3D point cloud with parametric surfaces. The method also comprises defining a partition of the parametric surfaces into oriented facets which respect intersections between the parametric surfaces. The method also comprises determining, among the oriented facets of the partition, a set of facets that represents a skin of the real object. The determining comprises minimizing an energy. The energy includes a data term and a constraint term. The data term increasingly penalizes discarding facets, as a level of fit between a discarded facet and the 3D point cloud increases. The constraint term penalizes formation of non-skin geometry by kept facets. Such a method provides an improved solution for 3D reconstruction.

公开(公告)号：US12190524B2

公开(公告)日：2025-01-07

申请号：US17384299

申请日：2021-07-23

Applicant: DASSAULT SYSTEMES

Inventor： Nicolas Beltrand ,  Mourad Boufarguine ,  Vincent Guitteny

IPC: G06T7/13 ,  G06N3/08 ,  G06T7/143 ,  G06T7/162 ,  G06T7/80 ,  G06T11/20

Abstract: A computer-implemented method for segmenting an object in at least one image acquired by a camera including computing an edge probabilities image based on the image, said edge probabilities image comprising, for each pixel of the image, the probability that said pixel is an edge, computing a segmentation probabilities image based on the image (IM), said segmentation probabilities image comprising, for each pixel of the image (IM), the probability that said pixel belongs to the object (OBJ), and computing a binary mask of the object based on the edge probabilities image and based on the segmentation probabilities image.

公开(公告)号：US10070121B2

公开(公告)日：2018-09-04

申请号：US14982591

申请日：2015-12-29

Applicant: DASSAULT SYSTEMES

Inventor： Mourad Boufarguine

IPC: G06T7/00 ,  H04N13/327 ,  H04N13/00 ,  H04N13/02 ,  H04N13/04 ,  H04N13/128 ,  H04N13/271 ,  G06T7/80 ,  G06T7/593

Abstract: A method for calibrating a depth camera comprising the steps of: S1) using a depth camera (DC) to be calibrated for acquiring a plurality of depth maps (DMS) of a physical three-dimensional reference object (PO), corresponding to different positions of the depth camera relative to the is reference object; S2) computing a three-dimensional reconstruction (3DRi) of the reference object from said depth maps, depending on values assigned to a set of calibration parameters (SCPi) of the depth camera; S3) computing an error (ERRi) between the three-dimensional reconstruction computed at step S2) and a predetermined digital model (RM) of the reference object; said steps S2) and S3) being iterated by assigning different values to said set of calibration parameters of the depth camera; and S4) choosing values (SCP) of said set of calibration parameters which minimize the error computed at step S3); said steps S2) to S4) being performed by a computer.