US 11,915,449 B2
Method and apparatus for estimating user pose using three-dimensional virtual space model
Nak Ju Doh, Seoul (KR); Ga Hyeon Lim, Seoul (KR); Jang Hun Hyeon, Seoul (KR); Dong Woo Kim, Seoul (KR); Bum Chul Jang, Seoul (KR); and Hyung A Choi, Seoul (KR)
Assigned to Korea University Research and Business Foundation, Seoul (KR)
Appl. No. 16/967,181
Filed by Korea University Research and Business Foundation, Seoul (KR)
PCT Filed Apr. 7, 2020, PCT No. PCT/KR2020/004685
§ 371(c)(1), (2) Date Aug. 4, 2020,
PCT Pub. No. WO2020/251151, PCT Pub. Date Dec. 17, 2020.
Claims priority of application No. 10-2019-0070829 (KR), filed on Jun. 14, 2019; and application No. 10-2020-0039564 (KR), filed on Apr. 1, 2020.
Prior Publication US 2023/0177723 A1, Jun. 8, 2023
Int. Cl. G06T 7/70 (2017.01); G06T 7/73 (2017.01); G06T 7/194 (2017.01)
CPC G06T 7/75 (2017.01) [G06T 7/194 (2017.01); G06T 2207/10028 (2013.01); G06T 2207/30196 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of estimating a user pose for a three-dimensional space, comprising:
a step of acquiring spatial information comprising depth information and image information for a three-dimensional space representing a real space using a depth measurement device and an image acquisition device;
a step of configuring depth-image associated information based on the spatial information and constructing a three-dimensional virtual space model corresponding to the real space and the three-dimensional space based on the depth-image associated information;
a step of receiving user information, comprising an image acquired by a user device, in the three-dimensional space;
a step of generating corresponding information corresponding to the user information in the three-dimensional virtual space model;
a step of calculating a similarity between the corresponding information and the user information; and
a step of estimating the user pose based on the similarity,
wherein the real space is a space, in which a discontinuous spatial map is generated when a spatial map is constructed using point cloud information,
wherein the real space comprises at least one of an indoor space, an outdoor space, and a space, in which the indoor space and outdoor space are linked, and
wherein the step of estimating the user pose comprises,
identifying a plurality of candidate poses and corresponding information for an entire area;
recalculating the similarity around a candidate pose of the plurality of candidate poses using the corresponding information;
repeating the recalculation for all of the plurality of candidate poses; and
estimating the user pose as one pose around the candidate pose with the highest similarity during the recalculation.