why sample t0 linearly from 0 to 1 in order to get a linear sampling in disparity

bmild / nerf

Code release for NeRF (Neural Radiance Fields)

http://tancik.com/nerf

MIT License

9.58k stars 1.34k forks source link

Open shuochen365 opened 3 years ago

shuochen365 commented 3 years ago

I can understand the paper moving the camera position origin to the z = -n plane.

t = -(near + rays_o[..., 2]) / rays_d[..., 2]

rays_o = rays_o + t[..., None] * rays_d

But I don't understand why t0 changes linearly from 0 to 1 in NDC space. Because according to the formula, it should be - 1 to 1.

NagabhushanSN95 commented 2 years ago

Check the document in this issue. o + tn * d is used to make it [0,1] in NDC space.