cv.stereoCalibrate() 的参数用于计算未失真图像的 R 和 T
我正在尝试校准两个不同的相机并获取外部数据。我已经使用cv.cameraCalibrate()来获取两个相机的内在函数，我想使用
cv.stereoCalibrate(objectPoints, imagePoints1, imagePoints2,cameraMatrix1, distCoeffs1,cameraMatrix2, distCoeffs2, imageSize )
获取 R 和 T 矩阵。但现在我不确定这个函数的参数，因为在找到棋盘之前我已经对图像进行了扭曲处理。我应该将 cameraMatrixes 和 distCoeffs 设置为什么值？
我尝试将它们设置为之前计算的矩阵，但这给出了完全错误的结果。如果我将它们设置为None，它们就在正确的范围内，但不太正确。现在我不确定什么是对或错，以及我之前是否犯了错误。
另外，由于 CALIB_FIX_INTRINSIC 标志已启用，我可以将 imageSize 设置为 None 吗？如果我正确理解文档，则不需要 imageSize ，因为我不想计算内在函数。
非常感谢 ：）
i'm trying to calibrate two different cameras and get the extrinsics. I've used cv.cameraCalibrate() to get the intrinsics of both cameras and i want to use
cv.stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize)
to get the R and T Matrix. But now i am unsure about the parameters to this function since i've undistorted my images before finding the chessboard. What values should i set the cameraMatrixes and the distCoeffs to?
I've tried to set them as the previously calculated matrixes, but that gives completely wrong results. If i set them to None they are in the right ballpark, but not quite right. Now i am unsure what's right or wrong and if i've made a mistake beforehand.
Also, can i set the imageSize to None, since the flag CALIB_FIX_INTRINSIC is enabled? If i understood the documentation correctly, then the imageSize isn't need, because i don't want to calculate the intrinsics.
Thank you very much :)
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论