abdallahdib

that seems too much, can u tell how many ray tracing samples are u using for optim?

so in fact the slow time are caused by these factors in ur case - texture resolution 2048 - env map 2048 - samples per pxiel (100) - running on...

well the final output is rendered with redner (saveOutput function in optimizer.py) this function render the face at high number of samples and save the outpout as an image (render.png)...

all the optimized textures are saved in png format with gamma correction applied to them

yes the function saveImage in Image.py already apply a gamma correction to the image. so u dont need to gamma correct the textures. u can (optionally) modify the code to...

Hi, NextFace uses cook torrance brdf for rendering, the estimated diffuse specular and roughness albedo maps can be used for this. u need to write ur own shader to combine...

also plz refer to the implementation of redner for accurate implmenetation of the brdf Line 355 (function bsdf) https://github.com/BachiLi/redner/blob/master/src/material.h

I m planning to replace redner with Mitsuba3. Any contribution is appreciated!

Hi Michael, U r talking about the deep version? it is put on hold currently but still planning to release it

Hi, Basically, **redner** is a differentiable ray tracing library written in C++ and **pyredner** is a python wrapper for redner. Did u follow the installation instructions in the INSTALL file...