公开(公告)号：US11244001B2

公开(公告)日：2022-02-08

申请号：US16724008

申请日：2019-12-20

Applicant: DASSAULT SYSTEMES

Inventor： Bastian Sdorra ,  Holger Dammertz

IPC: G06K9/46 ,  G06F16/583 ,  G06K9/62

Abstract: The disclosure notably relates to a computer-implemented method for retrieving a similar virtual material appearance from a database. The method comprises providing a database including virtual material appearances associated to an appearance signatures computed from values representing at least one of a structure, a reflection and a color of the associated virtual material appearance. The method further comprises providing a first appearance signature associated to a first virtual material appearance. The method further comprises identifying one or more virtual material appearances similar to the first virtual material appearance by comparing the first appearance signature with appearance signatures respectively associated to virtual material appearances in the database. The method improves retrieval of virtual material appearances.

公开(公告)号：US12094091B2

公开(公告)日：2024-09-17

申请号：US17557326

申请日：2021-12-21

Applicant: Dassault Systemes

Inventor： Andreas Weinmann ,  Holger Dammertz

IPC: G06T5/00 ,  G06N3/04 ,  G06N3/08 ,  G06T3/40 ,  G06T5/50 ,  G06T5/70

CPC classification number: G06T5/70 ,  G06N3/04 ,  G06N3/08 ,  G06T3/40 ,  G06T5/50 ,  G06T2207/20016 ,  G06T2207/20076 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20224

Abstract: A computer-implemented method of machine learning including learning a Convolutional Neural Network (CNN) architecture for estimating a degradation generated by a denoiser on a ray traced image. The method includes obtaining a dataset and learning the CNN architecture based on the obtained dataset. The learning including taking as input an image generated by the denoiser and a corresponding noisy image of the provided dataset and outputting an error map. This forms an improved solution with respect to estimating a degradation generated by a denoiser on a ray traced image.