US 12,430,872 B2
Method for detecting reflectivity information of deep learning-based virtual environment point cloud data, and electronic device performing same
Nam-il Kim, Seongnam-si (KR)
Assigned to NAVER LABS CORPORATION, Seongnam-si (KR)
Filed by NAVER LABS CORPORATION, Seongnam-si (KR)
Filed on Mar. 18, 2022, as Appl. No. 17/698,452.
Application 17/698,452 is a continuation of application No. PCT/KR2019/015240, filed on Nov. 11, 2019.
Claims priority of application No. 10-2019-0141344 (KR), filed on Nov. 7, 2019.
Prior Publication US 2022/0207858 A1, Jun. 30, 2022
Int. Cl. G06V 10/56 (2022.01); G06T 7/70 (2017.01); G06T 7/90 (2017.01); G06V 10/60 (2022.01); G06V 10/764 (2022.01); G06V 10/766 (2022.01); G06V 10/82 (2022.01); G06V 20/56 (2022.01)
CPC G06V 10/56 (2022.01) [G06T 7/70 (2017.01); G06T 7/90 (2017.01); G06V 10/60 (2022.01); G06V 10/764 (2022.01); G06V 10/766 (2022.01); G06V 10/82 (2022.01); G06V 20/56 (2022.01); G06T 2207/10024 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30252 (2013.01)] 17 Claims
OG exemplary drawing
 
1. An operating method of an electronic device, the method comprising:
acquiring image data and point cloud data;
detecting physical information that includes location information and color information about each point of the point cloud data based on the image data; and
detecting reflectivity information of the point cloud data by performing deep learning on the point cloud data based on the physical information; and
wherein the reflectivity information is estimated as a sum of class information about each of a plurality of classes classified at a predetermined distance and regression information of the rest thereof.