Title:
MODEL TRAINING METHOD AND SYSTEM FOR AUTOMATICALLY DETERMINING DAMAGE LEVEL OF EACH OF VEHICLE PARTS ON BASIS OF DEEP LEARNING
Document Type and Number:
WIPO Patent Application WO/2020/256246
Kind Code:
A1
Abstract:
The present invention relates to a model training method and system for automatically determining a damage level of each of vehicle parts on the basis of deep learning, wherein a model capable of rapidly calculating a coherent and reliable vehicle repair estimate is generated through training of a damage level according to damage types, and training allowing the model to automatically extract a photo through which a damage level can be determined, among photos of an accident vehicle by using a deep learning-based mask R-CNN framework and an inception V4 network structure.
Inventors:
KIM TAE YOUN (KR)
EO JIN SOL (KR)
BAE BYUNG SUN (KR)
EO JIN SOL (KR)
BAE BYUNG SUN (KR)
Application Number:
PCT/KR2019/018699
Publication Date:
December 24, 2020
Filing Date:
December 30, 2019
Export Citation:
Assignee:
AGILESODA INC (KR)
International Classes:
G06Q50/30; G06T5/00; G06T7/00
Foreign References:
| KR20180118596A | 2018-10-31 | |||
| KR20160018945A | 2016-02-18 | |||
| JP2002056122A | 2002-02-20 |
Other References:
"MASK R-CNN", 21 January 2018 (2018-01-21), XP055773410, Retrieved from the Internet [retrieved on 20200325]
"Representative CNN models-AlexN GoogLeNet, RestNet", BLOG.NAVER.COM, 23 December 2018 (2018-12-23), XP055773575, Retrieved from the Internet [retrieved on 20200325]
"Representative CNN models-AlexN GoogLeNet, RestNet", BLOG.NAVER.COM, 23 December 2018 (2018-12-23), XP055773575, Retrieved from the Internet
Attorney, Agent or Firm:
PARK, Min Heung et al. (KR)
Download PDF: