Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020069379 - TRAINING A DEEP NEURAL NETWORK MODEL TO GENERATE RICH OBJECT-CENTRIC EMBEDDINGS OF ROBOTIC VISION DATA

Publication Number WO/2020/069379
Publication Date 02.04.2020
International Application No. PCT/US2019/053554
International Filing Date 27.09.2019
IPC
G06K 9/00 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
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
G06K 9/72 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
72using context analysis based on the provisionally recognised identity of a number of successive patterns, e.g. a word
CPC
G06K 9/00664
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
00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
00664Recognising scenes such as could be captured by a camera operated by a pedestrian or robot, including objects at substantially different ranges from the camera
G06K 9/6262
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
6262Validation, performance evaluation or active pattern learning techniques
G06K 9/627
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
627based on distances between the pattern to be recognised and training or reference patterns
G06K 9/72
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
72using context analysis based on the provisionally recognised identity of a number of successive patterns, e.g. a word
Applicants
  • GOOGLE LLC [US]/[US]
Inventors
  • PIRK, Soeren
  • BAI, Yunfei
  • SERMANET, Pierre
  • KHANSARI ZADEH, Seyed Mohammad
  • LYNCH, Harrison
Agents
  • HIGDON, Scott
  • SALAZAR, John
  • SHUMAKER, Brantley
  • PURCELL, John
Priority Data
62/737,79427.09.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) TRAINING A DEEP NEURAL NETWORK MODEL TO GENERATE RICH OBJECT-CENTRIC EMBEDDINGS OF ROBOTIC VISION DATA
(FR) ENTRAÎNEMENT D'UN MODÈLE DE RÉSEAU NEURONAL PROFOND POUR GÉNÉRER DE RICHES INCORPORATIONS CENTRÉES SUR DES OBJETS DE DONNÉES DE VISION ROBOTIQUE
Abstract
(EN)
Training a machine learning model (e.g., a neural network model such as a convolutional neural network (CNN) model) so that, when trained, the model can be utilized in processing vision data (e.g., from a vision component of a robot), that captures an object, to generate a rich object-centric embedding for the vision data. The generated embedding can enable differentiation of even subtle variations of attributes of the object captured by the vision data.
(FR)
L'invention concerne l'entraînement d'un modèle d'apprentissage machine (par exemple, un modèle de réseau neuronal tel qu'un modèle de réseau neuronal convolutif (CNN)) de telle sorte que, lorsqu'il est entraîné, le modèle peut être utilisé dans le traitement des données de vision (par exemple, à partir d'un composant de vision d'un robot), qui capturent un objet, pour générer une riche incorporation centrée sur un objet pour les données de vision. L'incorporation générée peut permettre la différenciation de variations, même subtiles, d'attributs de l'objet capturé par les données de vision.
Latest bibliographic data on file with the International Bureau