Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020109497 - METHOD AND SYSTEM FOR GENERATING DATA

Publication Number WO/2020/109497
Publication Date 04.06.2020
International Application No. PCT/EP2019/082957
International Filing Date 28.11.2019
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 9/6247
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
6232Extracting features by transforming the feature space, e.g. multidimensional scaling; Mappings, e.g. subspace methods
6247based on an approximation criterion, e.g. principal component analysis
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
G06K 9/6278
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
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
6277based on a parametric (probabilistic) model, e.g. based on Neyman-Pearson lemma, likelihood ratio, Receiver Operating Characteristic [ROC] curve plotting a False Acceptance Rate [FAR] versus a False Reject Rate [FRR]
6278Bayesian classification
G06K 9/6298
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
6298Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
Applicants
  • PROWLER.IO LIMITED [GB]/[GB]
Inventors
  • HENSMAN, James
  • VAN DER WILK, Mark
  • DUTORDOIR, Vincent
Agents
  • EIP
Priority Data
18209633.930.11.2018EP
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) METHOD AND SYSTEM FOR GENERATING DATA
(FR) PROCÉDÉ ET SYSTÈME DE GÉNÉRATION DE DONNÉES
Abstract
(EN)
There is provided a computer-implemented method of training a computer system to generate output images in dependence on a plurality of training images. The method includes receiving training data corresponding to the plurality of training images. The method further includes initialising a first set of parameters comprising, for each of the plurality of training images, one or more parameters for a respective variational distribution over latent variables, and initialising a second set of parameters for a deep Gaussian process (GP). The deep GP comprises: a first GP defined by a first kernel and corresponding to a distribution over first functions for mapping latent variables to latent images, each latent image comprising a plurality of elements; and a second GP defined by a second kernel and corresponding to a distribution over second functions for mapping latent images to output images. The method further includes iteratively updating the first set of parameters and the second set of parameters to increase an average probability density associated with each training image being generated when a sampled latent variable is passed through a first function sampled from the first GP and a second function sampled from second GP. The second functions map patches of latent images to pixels of output images, each patch of a latent image comprising a subset of the elements of the latent image.
(FR)
L'invention concerne un procédé mis en oeuvre par ordinateur pour entraîner un système informatique à générer des images de sortie en fonction d'une pluralité d'images d'apprentissage. Le procédé consiste à recevoir des données d'apprentissage correspondant à la pluralité d'images d'apprentissage. Le procédé comprend en outre l'initialisation d'un premier ensemble de paramètres comprenant, pour chacune de la pluralité d'images d'apprentissage, un ou plusieurs paramètres pour une distribution variationnelle respective sur des variables latentes, et l'initialisation d'un second ensemble de paramètres pour un processus gaussien profond (GP). Le GP profond comprend: un premier GP défini par un premier noyau et correspondant à une distribution sur des premières fonctions pour mapper des variables latentes sur des images latentes, chaque image latente comprenant une pluralité d'éléments; et un second GP défini par un second noyau et correspondant à une distribution sur de secondes fonctions pour mapper des images latentes sur des images de sortie. Le procédé comprend en outre la mise à jour itérative du premier ensemble de paramètres et du second ensemble de paramètres pour augmenter une densité de probabilité moyenne associée à chaque image d'apprentissage générée lorsqu'une variable latente échantillonnée est soumise à travers une première fonction échantillonnée à partir de la première GP et une seconde fonction échantillonnée à partir de la seconde GP. Les secondes fonctions mappent des rustines d'images latentes sur des pixels d'images de sortie, chaque rustine d'une image latente comprenant un sous-ensemble des éléments de l'image latente.
Also published as
Latest bibliographic data on file with the International Bureau