SVBRDF estimation from photographs for three different lighting conditions (directional, natural, and flash/no-flash illumination) is shown by refining a novel SVBRDF diffusion backbone model.
Primary LanguageJupyter NotebookOtherNOASSERTION