Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

A question regarding visibility from the bvh raytracer #43

Open
swapb94 opened this issue Oct 13, 2024 · 0 comments
Open

A question regarding visibility from the bvh raytracer #43

swapb94 opened this issue Oct 13, 2024 · 0 comments

Comments

@swapb94
Copy link

swapb94 commented Oct 13, 2024

trace_results = raytracer.trace_visibility(
                rays_o,
                rays_d,
                means3D,
                cov_inv,
                opacity,
                normal)
visibility = trace_results['visibility']                

say, N: number of gaussians, S: number of sampled rays
From my understanding,

  • rays_o is [N, 3], which are the origins of each ray, i.e. each of the N Gaussian centers.
  • rays_d is [N, S, 3], which is essentially the direction of each of S rays that was fired from each of the N Gaussians.

The output visbility is [N, S, 1] with all binary values.
Q1. Is this telling us whether any Gaussian was hit for each of the fired ray?
Q2. If Q1 is true, then I would understand the visibility that we have as output above is the Gaussian-to-ray visibility. Any idea/hints on how to get the Gaussian-to-Gaussian visibility? basically a NxN matrix, which tells that, for any Gaussian point in the scene, which of the remainder N-1 points are "visible"
idea would be to culminate the hit result from all the sampled S rays for an individual point to get the equivalent row in the NxN matrix.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant