Inverse Rendering of Translucent Objects using Shape-adaptive Importance Sampling

POSTECH, Sogang University
Pacific Graphics 2024
Teaser Image

Inverse rendering with shape-adaptive importance sampling. (a-c) We jointly optimize the extinction coefficient and scattering albedo of a homo-geneous translucent object using the shape-BSSRDF model [Vicini 2019]. We address the challenges in differentiability posed by the importance sampling framework by approximating gradients using offset samples. (d, e) Our method computes highly accurate gradients with low variance.

Abstract

Subsurface scattering is ubiquitous in organic materials and has been widely researched in computer graphics. Inverse rendering of subsurface scattering, however, is often constrained by the planar geometry assumption of traditional analytic Bidirectional Scattering Surface Reflectance Distribution Functions (BSSRDF). To address this issue, a shape-adaptive BSSRDF model has been proposed to render translucent objects on curved geometry with high accuracy. In this paper, we leverage this model to estimate parameters of subsurface scattering for inverse rendering. We compute the finite difference of the rendering equation for subsurface scattering and iteratively update material parameters. We demonstrate the performance of our shape-adaptive inverse rendering model by analyzing the estimation accuracy and comparing to inverse rendering with plane-based BSSRDF models and volumetric methods.

Results

Method

Teaser Image

We analyzed challenges in differentiability of the shape-adaptive importance sampling framework due to 1) implicitly defined BSSRDFs and 2) non-invertibility of the sampling transform. Our solution is to compute gradients via the difference of offset samples, which is essentially finite differences in the sample domain. It is effective as it implicitly addresses boundary effects caused by parameter-dependent visibility discontinuities. It is also efficient becauses it utilizes the correlation of paths compared to naive automatic differentiation.

BibTeX

Coming Soon
l