Calibrating cameras on unmanned aerial vehicles using human joints

• Main Authors: BERESTOV, ALEXANDER, TAHARA DAISUKE
• Format: Patent
• Language: Chinese; English
• Published: 24.06.2022
• Subjects: CALCULATING; COMPUTING; CONTROLLING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; REGULATING; SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

Systems and methods for calibrating a camera on an unmanned aerial vehicle (UAV) are provided. The system receives a set of anchor images of a human subject from a set of anchor cameras and a set of images of the human subject from multiple viewpoints in a three-dimensional (3D) space from a camera group on a UAV group. The system determines a first set of two-dimensional (2D) positions of the human joint from the set of anchor images, and determines a second set of 2D positions of the human joint from the set of images. The system calculates 3D positions of the human joint as 3D keypoints based on triangulation using a first set of 2D positions and determines a 2D reprojection error between the 3D keypoints and a second set of 2D positions. Thereafter, the system calibrates each camera in the camera assembly by minimizing the 2D reprojection error. 提供了用于校准无人驾驶飞行器(UAV)上的摄影机的系统和方法。该系统从锚摄影机的集合接收人类被摄体的锚图像集合，并从UAV组上的摄影机组接收来自三维(3D)空间中多个视点的人类被摄体的图像组。该系统从锚图像集合确定人体关节的二维(2D)位置的第一集合，并从图像组确定人体关节的2D位置的第二集合。该系统使用2D位置的第一集合