A Simple Approach to Differentiable Rendering of SDFs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zichen, Deng, Xi, Zhang, Ziyi, Jakob, Wenzel, Marschner, Steve
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910476062949376
author Wang, Zichen
Deng, Xi
Zhang, Ziyi
Jakob, Wenzel
Marschner, Steve
author_facet Wang, Zichen
Deng, Xi
Zhang, Ziyi
Jakob, Wenzel
Marschner, Steve
contents We present a simple algorithm for differentiable rendering of surfaces represented by Signed Distance Fields (SDF), which makes it easy to integrate rendering into gradient-based optimization pipelines. To tackle visibility-related derivatives that make rendering non-differentiable, existing physically based differentiable rendering methods often rely on elaborate guiding data structures or reparameterization with a global impact on variance. In this article, we investigate an alternative that embraces nonzero bias in exchange for low variance and architectural simplicity. Our method expands the lower-dimensional boundary integral into a thin band that is easy to sample when the underlying surface is represented by an SDF. We demonstrate the performance and robustness of our formulation in end-to-end inverse rendering tasks, where it obtains results that are competitive with or superior to existing work.
format Preprint
id arxiv_https___arxiv_org_abs_2405_08733
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Simple Approach to Differentiable Rendering of SDFs
Wang, Zichen
Deng, Xi
Zhang, Ziyi
Jakob, Wenzel
Marschner, Steve
Graphics
Computer Vision and Pattern Recognition
We present a simple algorithm for differentiable rendering of surfaces represented by Signed Distance Fields (SDF), which makes it easy to integrate rendering into gradient-based optimization pipelines. To tackle visibility-related derivatives that make rendering non-differentiable, existing physically based differentiable rendering methods often rely on elaborate guiding data structures or reparameterization with a global impact on variance. In this article, we investigate an alternative that embraces nonzero bias in exchange for low variance and architectural simplicity. Our method expands the lower-dimensional boundary integral into a thin band that is easy to sample when the underlying surface is represented by an SDF. We demonstrate the performance and robustness of our formulation in end-to-end inverse rendering tasks, where it obtains results that are competitive with or superior to existing work.
title A Simple Approach to Differentiable Rendering of SDFs
topic Graphics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.08733