Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020142077 - METHOD AND SYSTEM FOR SEMANTIC SEGMENTATION INVOLVING MULTI-TASK CONVOLUTIONAL NEURAL NETWORK

Publication Number WO/2020/142077
Publication Date 09.07.2020
International Application No. PCT/US2018/068172
International Filing Date 31.12.2018
IPC
G06N 3/02 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
G06N 3/04 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
G06T 1/20 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
1General purpose image data processing
20Processor architectures; Processor configuration, e.g. pipelining
G06T 3/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
3Geometric image transformation in the plane of the image
G06T 7/246 2017.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
20Analysis of motion
246using feature-based methods, e.g. the tracking of corners or segments
CPC
G06N 3/0454
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
0454using a combination of multiple neural nets
G06N 3/084
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
084Back-propagation
G06T 2207/10016
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
10Image acquisition modality
10016Video; Image sequence
G06T 2207/20081
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
20Special algorithmic details
20081Training; Learning
G06T 2207/20084
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
20Special algorithmic details
20084Artificial neural networks [ANN]
G06T 2207/30261
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
2207Indexing scheme for image analysis or image enhancement
30Subject of image; Context of image processing
30248Vehicle exterior or interior
30252Vehicle exterior; Vicinity of vehicle
30261Obstacle
Applicants
  • BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO. LTD. [CN]/[CN]
Inventors
  • RUXIAO, Bao
  • XUN, Xu
Agents
  • EUREK, Justin
Priority Data
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) METHOD AND SYSTEM FOR SEMANTIC SEGMENTATION INVOLVING MULTI-TASK CONVOLUTIONAL NEURAL NETWORK
(FR) PROCÉDÉ ET SYSTÈME DE SEGMENTATION SÉMANTIQUE CONCERNANT UN RÉSEAU NEURONAL CONVOLUTIF
Abstract
(EN)
Methods and systems involving convolutional neural networks as applicable for semantic segmentation, including multi-task convolutional networks employing curriculum based transfer learning, are disclosed herein. In one example embodiment, a method of semantic segmentation involving a convolutional neural network includes training and applying the convolutional neural network. The training of the convolutional neural network includes each of training a semantic segmentation decoder network of the convolutional neural network, generating first feature maps by way of an encoder network of the convolutional neural network, based at least in part upon a dataset received at the encoder network, and training an instance segmentation decoder network of the convolutional neural network based at least in part upon the first feature maps. The applying includes receiving an image, and generating each of a semantic segmentation map and an instance segmentation map in response to the receiving of the image, in a single feedforward pass.
(FR)
La présente invention porte sur des procédés et des systèmes concernant des réseaux neuronaux convolutifs applicables à une segmentation sémantique y compris des réseaux convolutifs multitâches qui utilisent un apprentissage par transfert fondé sur un programme d'enseignement. Dans un mode de réalisation donné à titre d'exemple, un procédé de segmentation sémantique concernant un réseau neuronal convolutif comprend l'apprentissage et l'application du réseau neuronal convolutif. L'apprentissage du réseau neuronal convolutif comprend chacun des aspects suivants : l'apprentissage d'un réseau de décodeur de segmentation sémantique du réseau neuronal convolutif, la génération de premières cartes de caractéristiques au moyen d'un réseau d'encodeur du réseau neuronal convolutif, sur la base, au moins en partie, d'un ensemble de données reçu sur le réseau d'encodeur, et à entraîner un réseau de décodeur de segmentation d'instance du réseau neuronal convolutif sur la base, au moins en partie, des premières cartes de caractéristiques. L'application consiste à recevoir une image et à générer chaque carte de segmentation sémantique et chaque carte de segmentation d'instance en réponse à la réception de l'image, dans un seul passage à propagation avant.
Also published as
Latest bibliographic data on file with the International Bureau