US 12,339,232 B2
Method and device for three-dimensional object scanning with invisible markers
Xinyuan Li, South Setauket, NY (US); Yu Ji, Palo Alto, CA (US); Yanchen Liu, Palo Alto, CA (US); Xiaochen Hu, Palo Alto, CA (US); and Changxi Zheng, Palo Alto, CA (US)
Assigned to TENCENT AMERICA LLC
Filed by Tencent America LLC, Palo Alto, CA (US)
Filed on Nov. 14, 2023, as Appl. No. 18/509,263.
Application 18/509,263 is a continuation of application No. 18/154,754, filed on Jan. 13, 2023, granted, now 11,860,098.
Prior Publication US 2024/0241055 A1, Jul. 18, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G01N 21/64 (2006.01)
CPC G01N 21/6456 (2013.01) [G01N 21/6428 (2013.01); G01N 2021/6441 (2013.01); G01N 2021/6495 (2013.01); G01N 2021/6497 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A three-dimensional (3D) object scanning method, comprising:
controlling, when a light source is off in a first period of a capturing cycle, a first camera device to capture a first image of a target object, the target object being placed with markers having different visibilities with and without light from the light source;
controlling, when the light source is on in a second period of the same capturing cycle, a second camera device to capture a second image of the target object, wherein the second image is used to determine 3D information of the target object; and
pairing the 3D information of the target object corresponding to the second image of the capturing cycle and the first image of the same capturing cycle to form a training sample for training a gesture recognition model.