I’m trying to do some applications with KinectFusion, which requires to transform the generated obj file coordinate to the camera coordinate. I think the obj file coordinate is the world coordinate.(I don’t know if this understanding is right.) So when I run the kinectfusion code, I record the WorldToCameraTransform matrix. But each time the matrix of the camera’s initial position is the unit matrix. That’s weird! It means that the world coordinate coincides with the camera coordinate. The blue icon below is the WorldToCameraTransform matrix I recorded.I expect both the rotation matrix and the translation vector have values.
So I would like to ask whether converting the obj file to the camera coordinate system is correct or not by using the WorldToCameraTransform matrix. Another question is why the WorldToCameraTransform matrix is a 4*4 unit matrix in the the camera’s initial position.
hi honghailiao, are you talking about c++ code here?
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.