Title:
SELF-SUPERVISED LEARNING-BASED THREE-DIMENSIONAL HUMAN POSTURE ESTIMATION METHOD USING MULTI-VIEW IMAGES
Document Type and Number:
WIPO Patent Application WO/2022/131390
Kind Code:
A1
Abstract:
Provided is a self-supervised learning-based three-dimensional human posture estimation method using multi-view images. A three-dimensional human posture estimation method according to an embodiment of the present invention uses a self-supervised learning method among deep learning techniques to restore a three-dimensional posture by using only an image, two-dimensional posture information, and camera parameters. Accordingly, it is possible to optimize a network by using only its own input data and two-dimensional posture labels without three-dimensional label data.
More Like This:
Inventors:
YOON JU HONG (KR)
PARK MIN GYU (KR)
CHANG IN HO (KR)
KIM JE WOO (KR)
PARK MIN GYU (KR)
CHANG IN HO (KR)
KIM JE WOO (KR)
Application Number:
PCT/KR2020/018365
Publication Date:
June 23, 2022
Filing Date:
December 15, 2020
Export Citation:
Assignee:
KOREA ELECTRONICS TECHNOLOGY (KR)
International Classes:
G06T7/20; G06N20/00; G06T7/70; H04N13/204
Foreign References:
| US20200084427A1 | 2020-03-12 | |||
| JP2019029021A | 2019-02-21 | |||
| KR20190088379A | 2019-07-26 | |||
| JP2019133658A | 2019-08-08 |
Other References:
CHANG, INHO; PARK, MIN-GYU; KIM, JAEWOO; YOON, JU HONG: "Self-Supervised 3D Human Pose Estimation Using Multi-View and Camera Parameter", PROCEEDINGS OF THE INSTITUTE OF ELECTRONICS AND INFORMATION ENGINEERS (IEIE) AUTUMN CONFERENCE 2020, 1 November 2020 (2020-11-01), Korea, pages 421 - 423, XP009537606
Attorney, Agent or Firm:
NAM, Choong Woo (KR)
Download PDF:
Previous Patent: FPGA DESIGN METHOD AND SYSTEM FOR DEEP LEARNING ALGORITHM
Next Patent: MASK COMPRISING SEPARABLE HOLDING STRAP
Next Patent: MASK COMPRISING SEPARABLE HOLDING STRAP