This repository was archived by the owner on Feb 11, 2025. It is now read-only.
This repository was archived by the owner on Feb 11, 2025. It is now read-only.
Von mises distribuiton in RefNeRF #150
Description
Why did you parameterize the reflection vector as a distribution, rather than using a single vector in Ref-NeRF.
'Next, we enable the directional MLP to reason about materials with different roughnesses by encoding a distribution of reflection vectors instead of a single vector' I saw this paragraph in your article but couldn't be able to get the intuition of why using a distribution to represent a single reflection vector. what is the advantage in it?