You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I want to: Create a scene with base_color as a differentiable UV texture, only optimize the red channel of this texture, while maintaining the computation graph for proper gradient flow
What I have tried: create a 2d array as a latent variable(representing red channel in UV space), apply a transformation where i create a new texture with G, B values from the orginal base_color texture map, and copy the red channel values to the new texture, but this is breaking the computation graph.
Question: Is there a proper way in Mitsuba 3 to optimize just one channel of a texture while maintaining the gradient flow through the computation graph? Perhaps using masks or some other technique that preserves differentiability?
Any examples or guidance would be greatly appreciated!
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
I want to: Create a scene with base_color as a differentiable UV texture, only optimize the red channel of this texture, while maintaining the computation graph for proper gradient flow
What I have tried: create a 2d array as a latent variable(representing red channel in UV space), apply a transformation where i create a new texture with G, B values from the orginal base_color texture map, and copy the red channel values to the new texture, but this is breaking the computation graph.
Question: Is there a proper way in Mitsuba 3 to optimize just one channel of a texture while maintaining the gradient flow through the computation graph? Perhaps using masks or some other technique that preserves differentiability?
Any examples or guidance would be greatly appreciated!
Beta Was this translation helpful? Give feedback.
All reactions