How to transform camera to up_right_depth coordinate ?

[phy12321]

Thanks for your great work!

I want to create my own dataset(I have rgbd only), so I turned depth.png into point cloud(x,y,z)， which is in camera coordinate system. But how to turn it in up_right coordinate system? I found reletive function in sungbrd_utils.py :

def project_upright_depth_to_camera(self, pc): pc2 = np.dot(np.transpose(self.Rtilt), np.transpose(pc[:,0:3])) # (3,n) return flip_axis_to_camera(np.transpose(pc2)) but how to get matrix Rtilt? (what this matrix means?) Look forward to your reply.

[alar0330]

Hi,

you don't actually have to convert the coordinates to upright, as long as you are consistent with the choice of your coordinate system throughout the entire model input.

[lyp0413]

@alar0330

you don't actually have to convert the coordinates to upright, as long as you are consistent with the choice of your coordinate system throughout the entire model input.

Hello, If I use pointclouds captured from an irregularly moving camera,can it predict the 6Dof infomation?

[rXYZkit]

I think that after you use rgb+depth to calculate the coordinates of the points in the point cloud (in the camera coordinate system), you need to convert these points to the world coordinate system. Here the world coordinate system is the upright-depth coordinate system. As alar0330 said, it should be feasible if it can be consistent in all places (for example, 8 corner coordinates, object center coordinates, etc.)