Title:
METHOD OF TRAINING A DEVICE USING MACHINE LEARNING
Document Type and Number:
WIPO Patent Application WO/2022/050678
Kind Code:
A2
Abstract:
An object of the present disclosure is to provide a pluggable camera that is mounted on top of shelf units can be equipped with AI and embedded computer vision algorithms to identify shoppers and monitor their shopping activities. To this end, the camera provides an easy to install and scalable solution that can detect items being picked up or returned to shelves by a shopper for tracking shopper behavior analytics and inventory management systems. The pluggable camera is a highly scalable solution because the camera can be mounted onto any unit shelf for an easy installation in existing store fronts such that it would require minimal to no change in the layout of the stores and no need for external calibration of the cameras.
Inventors:
LEE BAISUB (US)
SONG JAIGAK (US)
SUPE BHOOSHAN (US)
GUACK JUNGICK (US)
SILVA HELDER (US)
KHALILI AMIR HOSSEIN (US)
PATEL SHANTANU (US)
HUYNH JULIE (US)
SARAF GAURAV (US)
SONG JAIGAK (US)
SUPE BHOOSHAN (US)
GUACK JUNGICK (US)
SILVA HELDER (US)
KHALILI AMIR HOSSEIN (US)
PATEL SHANTANU (US)
HUYNH JULIE (US)
SARAF GAURAV (US)
Application Number:
PCT/KR2021/011748
Publication Date:
March 10, 2022
Filing Date:
September 01, 2021
Export Citation:
Assignee:
LG ELECTRONICS INC (KR)
International Classes:
G06N20/00
Attorney, Agent or Firm:
KBK & ASSOCIATES (KR)
Download PDF: