Some content of this application is unavailable at the moment.
If this situation persist, please contact us atFeedback&Contact
1. (WO2019050247) NEURAL NETWORK LEARNING METHOD AND DEVICE FOR RECOGNIZING CLASS
Latest bibliographic data on file with the International Bureau    Submit observation

Pub. No.: WO/2019/050247 International Application No.: PCT/KR2018/010271
Publication Date: 14.03.2019 International Filing Date: 04.09.2018
IPC:
G06N 3/08 (2006.01) ,G06N 3/04 (2006.01)
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
08
Learning methods
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
04
Architecture, e.g. interconnection topology
Applicants:
삼성전자 주식회사 SAMSUNG ELECTRONICS CO., LTD. [KR/KR]; 경기도 수원시 영통구 삼성로 129 129, Samsung-ro, Yeongtong-gu, Suwon-si, Gyeonggi-do 16677, KR
한국과학기술원 KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY [KR/KR]; 대전시 유성구 대학로 291 291, Daehak-ro, Yuseong-gu, Daejeon 34141, KR
Inventors:
김준모 KIM, Jun-mo; KR
김병주 KIM, Byung-ju; KR
김주창 KIM, Joo-chang; KR
이예강 LEE, Ye-gang; KR
박민석 PARK, Min-seok; KR
윤주승 YUN, Ju-seung; KR
이재영 LEE, Jae-young; KR
주동규 JOO, Dong-gyu; KR
Agent:
리앤목 특허법인 Y.P.LEE, MOCK & PARTNERS; 서울시 강남구 언주로30길 13 대림아크로텔 12층 12F Daelim Acrotel 13 Eonju-ro 30-gil, Gangnam-gu, Seoul 06292, KR
Priority Data:
10-2017-011545108.09.2017KR
Title (EN) NEURAL NETWORK LEARNING METHOD AND DEVICE FOR RECOGNIZING CLASS
(FR) PROCÉDÉ ET DISPOSITIF D'APPRENTISSAGE DE RÉSEAU DE NEURONES ARTIFICIELS POUR RECONNAÎTRE UNE CLASSE
(KO) 클래스 인식을 위한 뉴럴 네트워크 학습 방법 및 디바이스
Abstract:
(EN) The present disclosure relates to a neural network learning method for recognizing the class of an object included in an image on the basis of an artificial intelligence system and an application thereof. The method enables: by using a first learned network model learned on the basis of source learning images respectively included in at least one class, acquiring property information of a query image included in a class different from the at least one class; acquiring a generated image from the property information of the query image by using a second learned network model; acquiring property information of the acquired generated image by using the first learned network model; and updating the respective weighted values of layers respectively included in the first learned network model and the second learned network model on the basis of the difference between the property information of the query image and the property information of the generated image, and the difference between the query image and the generated image.
(FR) La présente invention concerne un procédé d'apprentissage de réseau de neurones artificiels pour reconnaître la classe d'un objet inclus dans une image sur la base d'un système d'intelligence artificielle et une application de celui-ci. Le procédé permet : à l'aide d'un premier modèle de réseau appris qui a été appris sur la base d'images d'apprentissage source incluses respectivement dans au moins une classe, d'acquérir des informations de propriété d'une image d'interrogation incluse dans une classe différente de ladite classe ; d'acquérir une image générée à partir des informations de propriété de l'image d'interrogation à l'aide d'un second modèle de réseau appris ; d'acquérir des informations de propriété de l'image générée acquise à l'aide du premier modèle de réseau appris ; et de mettre à jour les valeurs pondérées respectives de couches incluses respectivement dans le premier modèle de réseau appris et le second modèle de réseau appris sur la base de la différence entre les informations de propriété de l'image d'interrogation et les informations de propriété de l'image générée, et de la différence entre l'image d'interrogation et l'image générée.
(KO) 본 개시는 인공지능 시스템 및 그 응용에 따라 이미지에 포함된 객체의 클래스 인식을 위한 뉴럴 네트워크 학습 방법으로, 적어도 하나의 클래스 각각에 포함된 소스 학습 이미지를 기초로 학습된 제 1 학습 네트워크 모델을 이용하여, 적어도 하나의 클래스와 다른 클래스에 포함된 쿼리(query) 이미지의 특성 정보를 획득하고, 제 2 학습 네트워크 모델을 이용하여, 쿼리 이미지의 특성 정보로부터 생성 이미지를 획득하며, 제 1 학습 네트워크 모델을 이용하여, 획득된 생성 이미지의 특성 정보를 획득하고, 쿼리 이미지의 특성 정보와 생성 이미지의 특성 정보 간의 차이 및 쿼리 이미지와 생성 이미지 간의 차이에 기초하여, 제 1 학습 네트워크 모델 및 제 2 학습 네트워크 모델 각각에 포함된 레이어의 가중치를 업데이트 할 수 있다.
front page image
Designated States: AE, AG, AL, AM, AO, AT, AU, AZ, BA, BB, BG, BH, BN, BR, BW, BY, BZ, CA, CH, CL, CN, CO, CR, CU, CZ, DE, DJ, DK, DM, DO, DZ, EC, EE, EG, ES, FI, GB, GD, GE, GH, GM, GT, HN, HR, HU, ID, IL, IN, IR, IS, JO, JP, KE, KG, KH, KN, KP, KW, KZ, LA, LC, LK, LR, LS, LU, LY, MA, MD, ME, MG, MK, MN, MW, MX, MY, MZ, NA, NG, NI, NO, NZ, OM, PA, PE, PG, PH, PL, PT, QA, RO, RS, RU, RW, SA, SC, SD, SE, SG, SK, SL, SM, ST, SV, SY, TH, TJ, TM, TN, TR, TT, TZ, UA, UG, US, UZ, VC, VN, ZA, ZM, ZW
African Regional Intellectual Property Organization (ARIPO) (BW, GH, GM, KE, LR, LS, MW, MZ, NA, RW, SD, SL, ST, SZ, TZ, UG, ZM, ZW)
Eurasian Patent Office (AM, AZ, BY, KG, KZ, RU, TJ, TM)
European Patent Office (EPO) (AL, AT, BE, BG, CH, CY, CZ, DE, DK, EE, ES, FI, FR, GB, GR, HR, HU, IE, IS, IT, LT, LU, LV, MC, MK, MT, NL, NO, PL, PT, RO, RS, SE, SI, SK, SM, TR)
African Intellectual Property Organization (BF, BJ, CF, CG, CI, CM, GA, GN, GQ, GW, KM, ML, MR, NE, SN, TD, TG)
Publication Language: Korean (KO)
Filing Language: Korean (KO)