Explanation regarding camera parameters #110
Description
I was trying out the deform demo, and my goal is to load custom silhouettes and transformation matrices of an object exported from perhaps a software like colmap and learn mesh from it.
I got stuck trying to understand the 4 lines following this line in the code
SoftRas/examples/demo_deform.py
Line 87 in b3150cf
Can someone please explain what are these variables with respect to the standard K[R|T] matrix (intrinsic [ extrinsic rotation | extrinsic translation] )
camera_distances = torch.from_numpy(cameras[:, 0])
elevations = torch.from_numpy(cameras[:, 1])
viewpoints = torch.from_numpy(cameras[:, 2])
Any help would be greatly appreciated !
Thanks