Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020112189 - COMPUTER ARCHITECTURE FOR ARTIFICIAL IMAGE GENERATION USING AUTO-ENCODER

Publication Number WO/2020/112189
Publication Date 04.06.2020
International Application No. PCT/US2019/047993
International Filing Date 23.08.2019
IPC
G06K 9/62 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
CPC
G06K 9/00201
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00201Recognising three-dimensional objects, e.g. using range or tactile information
G06K 9/6256
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
G06N 20/10
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
10using kernel methods, e.g. support vector machines [SVM]
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
G06T 15/60
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
153D [Three Dimensional] image rendering
50Lighting effects
60Shadow generation
G06T 17/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
17Three dimensional [3D] modelling, e.g. data description of 3D objects
Applicants
  • RAYTHEON COMPANY [US]/[US]
Inventors
  • KIM, Peter
  • SAND, Michael J.
  • HOLLENBECK, Matthew D.
Agents
  • PERDOK, Monique M.
  • ARORA, Suneel, Reg. No. 42,267
  • BEEKMAN, Marvin L., Reg. No. 38,377
  • BIANCHI, Timothy E., Reg. No. 39,610
  • BLACK, David W., Reg. No. 42,331
  • LANG, Allen R., Reg. No. 58,829
  • MCCRACKIN, Ann M., Reg. No. 42,858
  • SCHEER, Bradley W., Reg. No. 47,059
Priority Data
62/771,80827.11.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) COMPUTER ARCHITECTURE FOR ARTIFICIAL IMAGE GENERATION USING AUTO-ENCODER
(FR) ARCHITECTURE D'ORDINATEUR POUR LA GÉNÉRATION D'IMAGES ARTIFICIELLES À L'AIDE D'UN AUTOENCODEUR
Abstract
(EN)
A computer architecture for artificial image generation is disclosed. According to some aspects, a computing machine receives a voxel model of a first set of objects different from a target object. The target object is to be recognized using an image recognizer. The computing machine generates, based on the voxel model, a set of TSB (target shadow background-mask) images of the first set of objects. The computing machine receives, at an auto-encoder, a set of real images of the first set of objects. The computing machine generates, using the auto-encoder, one or more artificial images of the target object based on the set of TSB images. The auto-encoder learns differences between the target object and the first set of objects. The computing machine provides, as output, the generated one or more artificial images of the target object.
(FR)
L'invention concerne une architecture d'ordinateur pour la génération d'images artificielles. Selon certains aspects, une machine de calcul reçoit un modèle à voxels d'un premier ensemble d'objets différents d'un objet cible. L'objet cible doit être reconnu à l'aide d'un système de reconnaissance d'image. La machine de calcul génère, sur la base du modèle à voxels, un ensemble d'images TSB (de l'anglais "target shadow background-mask", masque cible/ombre/arrière-plan) du premier ensemble d'objets. La machine de calcul reçoit, au niveau d'un autoencodeur, un ensemble d'images réelles du premier ensemble d'objets. La machine de calcul génère, à l'aide de l'autoencodeur, une ou plusieurs images artificielles de l'objet cible sur la base de l'ensemble d'images TSB. L'autoencodeur apprend les différences entre l'objet cible et le premier ensemble d'objets. La machine de calcul fournit, en sortie, la ou les images artificielles générées de l'objet cible.
Latest bibliographic data on file with the International Bureau