The reviewed record of science sign in
Pith

arxiv: 2111.00140 · v1 · pith:AGEWQTH4 · submitted 2021-10-30 · cs.CV · cs.GR

DIB-R++: Learning to Predict Lighting and Material with a Hybrid Differentiable Renderer

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:AGEWQTH4record.jsonopen to challenge →

classification cs.CV cs.GR
keywords lightingmaterialdifferentiablerendererrenderersapproachescompareddibr
0
0 comments X
read the original abstract

We consider the challenging problem of predicting intrinsic object properties from a single image by exploiting differentiable renderers. Many previous learning-based approaches for inverse graphics adopt rasterization-based renderers and assume naive lighting and material models, which often fail to account for non-Lambertian, specular reflections commonly observed in the wild. In this work, we propose DIBR++, a hybrid differentiable renderer which supports these photorealistic effects by combining rasterization and ray-tracing, taking the advantage of their respective strengths -- speed and realism. Our renderer incorporates environmental lighting and spatially-varying material models to efficiently approximate light transport, either through direct estimation or via spherical basis functions. Compared to more advanced physics-based differentiable renderers leveraging path tracing, DIBR++ is highly performant due to its compact and expressive shading model, which enables easy integration with learning frameworks for geometry, reflectance and lighting prediction from a single image without requiring any ground-truth. We experimentally demonstrate that our approach achieves superior material and lighting disentanglement on synthetic and real data compared to existing rasterization-based approaches and showcase several artistic applications including material editing and relighting.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.