US 12,406,379 B2
Method and apparatus for detecting motion information of target, device and medium
Wenming Meng, Zhejiang (CN); Hongmei Zhu, Zhejiang (CN); and Qian Zhang, Zhejiang (CN)
Assigned to Horizon Journey (Hangzhou) Artificial Intelligence Technology Co., Ltd., Zhejiang (CN)
Appl. No. 17/907,662
Filed by Horizon Journey (Hangzhou) Artificial Intelligence Technology Co., Ltd., Zhejiang (CN)
PCT Filed Feb. 18, 2022, PCT No. PCT/CN2022/076765
§ 371(c)(1), (2) Date Sep. 28, 2022,
PCT Pub. No. WO2022/213729, PCT Pub. Date Oct. 13, 2022.
Claims priority of application No. 202110373003.X (CN), filed on Apr. 7, 2021.
Prior Publication US 2024/0212170 A1, Jun. 27, 2024
Int. Cl. G06T 7/246 (2017.01); G05D 1/243 (2024.01); G06T 7/55 (2017.01); G06T 7/73 (2017.01)
CPC G06T 7/246 (2017.01) [G05D 1/2435 (2024.01); G05D 1/2437 (2024.01); G06T 7/55 (2017.01); G06T 7/73 (2017.01); G06T 2207/10016 (2013.01); G06T 2207/30252 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting motion information of a target, comprising:
performing target detection on a first image to obtain a detection box of a first target, wherein the first image is an image of a scene outside a traveling object that is captured by an image capturing device on the traveling object in a traveling process of the traveling object;
acquiring depth information of the first image in a corresponding first camera coordinate system;
determining depth information of the detection box of the first target based on the depth information of the first image in the corresponding first camera coordinate system, and determining first coordinates of the first target in the first camera coordinate system based on a location of the detection box of the first target in an image coordinate system and the depth information of the detection box of the first target;
acquiring pose change information of the image capturing device from capturing of a second image to capturing of the first image, wherein the second image is an image that is before the first image in terms of timing and spaced apart from the first image by a preset number of frames in an image sequence where the first image is present;
transforming second coordinates of a second target in a second camera coordinate system corresponding to the second image into third coordinates in the first camera coordinate system based on the pose change information, wherein the second target is a target in the second image that corresponds to the first target; and
determining motion information of the first target within a corresponding time range from a capturing time point of the second image to that of the first image based on the first coordinates and the third coordinates.