CONVOLUTIONAL NEURAL NETWORK STRUCTURE FOR EXPLAINABLE 3D SHAPE LEARNING, AND METHOD AND SYSTEM FOR 3D SHAPE LEARNING USING SAME

Title:

CONVOLUTIONAL NEURAL NETWORK STRUCTURE FOR EXPLAINABLE 3D SHAPE LEARNING, AND METHOD AND SYSTEM FOR 3D SHAPE LEARNING USING SAME

Document Type and Number:

WIPO Patent Application WO/2024/035000

Kind Code:

A1

Abstract:

Disclosed are a convolutional neural network structure for explainable 3D shape learning, and a method and a system for 3D shape learning by using same. A method for shape learning performed by a shape learning system according to an embodiment may comprise the steps of: extracting a geodetic feature and a geometric feature from three-dimensional shape data; and classifying an object via a convolution operation on the extracted geodetic feature and the extracted geometric feature with respect to planes constituting the three-dimensional shape data.

More Like This:

JP7446414	Method and related apparatus for obtaining textures of 3D models
WO/2024/116059	DISPLAY SYSTEM, DISPLAY METHOD, AND RECORDING MEDIUM
WO/2022/187574	SYSTEM AND METHOD FOR TRACKING AN OBJECT BASED ON SKIN IMAGES

Inventors:

CHAE DONG-KYU (KR)
KIM SEONGGYEOM (KR)

Application Number:

PCT/KR2023/011438

Publication Date:

February 15, 2024

Filing Date:

August 03, 2023

Export Citation:

Click for automatic bibliography generation Help

Assignee:

IUCF HYU (KR)

International Classes:

G06V20/64; G06N3/08; G06T7/60; G06V10/422; G06V10/764; G06V10/82

Foreign References:

KR20180126220A	2018-11-27
KR20190050639A	2019-05-13
KR101007276B1	2011-01-13
KR20210136597A	2021-11-17
KR102157793B1	2020-09-18

Attorney, Agent or Firm:

YANG, Sungbo (KR)

Download PDF:

View/Download PDF PDF Help

Previous Patent: METHOD AND DEVICE FOR PROVIDING SCENARIO FOR SURGERY

Next Patent: DEVICE FOR MANUFACTURING TEXTURED VEGETABLE PROTEIN