Help with implementing fluorescent material

[arpit15]

I want to implement a basic fluorescent material given a re-radiation matrix.
First idea is to expand the BSDFContext struct to add L_o or L_i based on the transport mode. Then, use this spectrum information to write eval, sample and pdf functions based on re-radiation matrix. However, this requires to change the integrator code to take the processed radiance after interaction with this material instead of multiplying the returned value.

Second idea is to write a custom class which outputs a re-radiation matrix, similar to polarization BSDFs, and write custom multiplication operators for obtain the radiance value after interaction with the material.

Any help or pointer on how to implement fluorescent material is welcome.
I am aware of ART renderer. I will use it to test the accuracy of the implemented material.

[njroussel]

Hi @arpit15

The "cheap" way is to just modify si.wavelenghts wherever needed. However, proper importance sampling would require some changes to the interfaces. I haven't really put much thought into it, but your first suggestion seems to be the more natural interface.