weiyithu/NerfingMVS

Question regarding guided sampling

melhashash opened this issue · 1 comments

melhashash commented

Hi, thanks for open-sourcing the project

I have a question regarding these two lines:

NerfingMVS/src/run_nerf.py

Line 373 in 5890f15

near = (depth_priors * (1 - torch.clamp(depth_confidences, min=near_bound, max=far_bound))).unsqueeze(1)

NerfingMVS/src/run_nerf.py

Line 374 in 5890f15

far = (depth_priors * (1 + torch.clamp(depth_confidences, min=near_bound, max=far_bound))).unsqueeze(1)

I think you did not use the near_bound and far_bound as in equation (6) in the paper, is it correct, or I am missing something?

gjgjgjfff commented

Hi, thanks for open-sourcing the project

I have a question regarding these two lines:

NerfingMVS/src/run_nerf.py

Line 373 in 5890f15

near = (depth_priors * (1 - torch.clamp(depth_confidences, min=near_bound, max=far_bound))).unsqueeze(1)

NerfingMVS/src/run_nerf.py

Line 374 in 5890f15

far = (depth_priors * (1 + torch.clamp(depth_confidences, min=near_bound, max=far_bound))).unsqueeze(1)

I think you did not use the near_bound and far_bound as in equation (6) in the paper, is it correct, or I am missing something?

Hello, I have the same question, why is near_bound and far_bound used here？and how should these two values be set if it is my own data set?