search

ankitdhall / lidar_camera_calibration

ROS package to find a rigid-body transformation between a LiDAR and a camera for "LiDAR-Camera Calibration using 3D-3D Point correspondences"
http://arxiv.org/abs/1705.09785
GNU General Public License v3.0
1.48k stars 459 forks source link

how to obtain 3D points in the camera frame #60

Open xccoming opened 5 years ago

xccoming commented 5 years ago

how to obtain 3D points in the camera frame, for mono camera, in real scale ? in paper: "The tags provide [R|t] between the camera and the center of the marker. This transform can be used to convert corner points from the marker’s frame-ofreference (which is the cardboard plane with the origin being the center of the ArUcO marker) to the camera’s frame-of-reference. This allows to obtain the corners as 3D points in the camera frame. " but this t is not in real scale!

Chrislzy1993 commented 5 years ago

@xccoming , i am confused too, did u solve this problem.

icameling commented 5 years ago

从aruco_mapping 得到tag到camera的转换矩阵 (R|t) ，这里单位是米 由 /conf/marker_coordinates.txt 输入标定板尺寸，得到标定板四个角点在tag坐标系下的位置，单位也是米 将在tag坐标系下的四个角点经Rt转换到相机坐标系，得到在相机坐标系下4个角点的空间3D位置。 两个标定板就是8个点。

对于激光则用直线拟合的方法得到这8个角点位置。

于是问题变成了： 已知两组对应的8D点，求两组点之间的转换关系。用Least-Squares Rigid Motion Using SVD 论文里面提出方法即可解决，也就是本论文里面的4.1部分。

Chrislzy1993 commented 5 years ago

@icameling , 谢谢你， 请问下对于marker和tag的尺寸应该怎样确定，还有对于单目相机，标定的效果怎么样？

xinwf commented 5 years ago

@icameling 我运行这个之后卡在了回调函数前面，怎么回事呢，输出"done1“之后就没有任何输出了，也没有弹出窗口，topic都是有数据的。

narutojxl commented 4 years ago

Hi @icameling, how to get 3D points corners of calibration board in laser frame, details see #107, thanks a lot!

SinghiDivyanshu commented 4 years ago

hi @xccoming , Did you solve it. I'm not able to obtain 3D points in the camera frame, for mono camera.