US 11,704,833 B2
Monocular vision tracking method, apparatus and non-transitory computer-readable storage medium
Yuan Lin, Beijing (CN); Yongjie Zhang, Beijing (CN); Yingchao Li, Beijing (CN); and Zhongqin Wu, Beijing (CN)
Assigned to Baidu Online Network Technology (Beijing) Co., Ltd., Beijing (CN)
Filed by Baidu Online Network Technology (Beijing) Co., Ltd, Beijing (CN)
Filed on Apr. 15, 2021, as Appl. No. 17/231,796.
Application 17/231,796 is a continuation of application No. 16/265,350, filed on Feb. 1, 2019, granted, now 10,984,554.
Claims priority of application No. 201810539448.9 (CN), filed on May 30, 2018.
Prior Publication US 2021/0233275 A1, Jul. 29, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/80 (2017.01); G06T 7/70 (2017.01)
CPC G06T 7/80 (2017.01) [G06T 7/70 (2017.01); G06T 2207/30244 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A monocular vision tracking method, comprising:
acquiring a first camera pose by using a wireless tracker;
capturing calibration board images by using a camera, and selecting a set of images from the captured calibration board images according to image sharpness and camera pose difference;
calculating a second camera pose in the camera calibration algorithm according to the selected set of images;
obtaining a conversion parameter between the first camera pose and the second camera pose; and
converting a first capturing pose into a second capturing pose by means of the conversion parameter when a scenario is captured, wherein the first capturing pose is acquired by the wireless tracker, and the second capturing pose corresponds to a pose of the camera,
wherein a relative position between the wireless tracker and the camera remains unchanged, in processes of acquiring the first camera pose, capturing the calibration board images, and capturing the scenario.