Gradient for CUDA point sampling from rays

sunset1995 / DirectVoxGO

Direct voxel grid optimization for fast radiance field reconstruction.

https://sunset1995.github.io/dvgo

Other

1.05k stars 110 forks source link

Gradient for CUDA point sampling from rays #20

Open PNeigel opened 2 years ago

PNeigel commented 2 years ago

Hi,

I've got a question regarding the CUDA extensions, more precise the following function:

https://github.com/sunset1995/DirectVoxGO/blob/9b172bfc9f0d349fb4b9db59caecd77b99ec7465/lib/cuda/render_utils_kernel.cu#L190

How difficult would it be to add a gradient for this function so that pytorch's autograd can backpropagate trough?

sunset1995 commented 2 years ago

Let me try it.

yashbhalgat commented 1 year ago

Hi @sunset1995 @PNeigel, is there any update on this issue? I am trying to perform NeRF-- style pose refinement and I need to backpropagate through this function too

wrld commented 1 year ago

Is there any update? I met the same problem.