3D METHOD AND APPARTUS FOR LEARNING-BASED GENERATING 3D MODEL

According to the present invention, provided is an operating method of a learning-based 3D model generating device, which includes the steps of: generating a multi-view feature image using supervised learning; generating a 3D mesh model using a point cloud corresponding to the multi-view feature ima...

Full description

Saved in:
Bibliographic Details
Main Authors CHOI JIN SUNG, KIM TAE JOON, HWANG BON WOO, LEE SEUNG WOOK, YOON SEUNG UK, LIM SEONG JAE
Format Patent
LanguageEnglish
Korean
Published 29.09.2021
Subjects
CALCULATING
COMPUTING
COUNTING
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:According to the present invention, provided is an operating method of a learning-based 3D model generating device, which includes the steps of: generating a multi-view feature image using supervised learning; generating a 3D mesh model using a point cloud corresponding to the multi-view feature image and the feature image representing internal shape information; generating a texture map by projecting the 3D mesh model onto the input images of three viewpoints; and generating a 3D model using the texture map. 본 발명에 따른 학습기반 3D 모델 생성 장치의 동작 방법은, 감독 학습을 이용하여 다시점 특성 이미지를 생성하는 단계, 상기 다시점 특성 이미지에 대응하는 포인트 클라우드와 내부 형태 정보를 표현하는 특성 이미지를 이용하여 3D 메쉬 모델을 생성하는 단계, 상기 3D 메쉬 모델을 입력된 세 시점의 이미지들에 투영함으로써 텍스처 맵을 생성하는 단계, 및 상기 텍스처 맵을 이용하여 3D 모델을 생성하는 단계를 포함할 수 있다
Bibliography:Application Number: KR20210121652