saeid-h/bts-fully-tf

Question about information required for training other than RGB Depth data

mohammad-saber opened this issue · 3 comments

Thank you very much for sharing your great code. I have a question about training process.

Is there any additional information other than RGB and depth required for training? (Example: camera intrinsic, extrinsic, or other project specific information).

Thank you very much for your help.

You need the camera's focal length.

May I ask another question in this regard?
What is purpose of using "camera's focal length"?

They use it to project 3D points on the 2D plane. The related paper is available here: https://arxiv.org/pdf/1907.10326.pdf
If you have more questions please feel free to contact me or the author by email.