Unbiased Gradient Estimation for Differentiable Surface Splatting via Poisson Sampling

Conference Paper (2022)
Author(s)

Jan Müller (Universität Bonn)

M. Weinmann (TU Delft - Computer Graphics and Visualisation)

Reinhard Klein (Universität Bonn)

Research Group
Computer Graphics and Visualisation
Copyright
© 2022 Jan U. Müller, M. Weinmann, Reinhard Klein
DOI related publication
https://doi.org/10.1007/978-3-031-19827-4_17
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 Jan U. Müller, M. Weinmann, Reinhard Klein
Research Group
Computer Graphics and Visualisation
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. @en
Pages (from-to)
281-299
ISBN (print)
9783031198267
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We propose an efficient and GPU-accelerated sampling framework which enables unbiased gradient approximation for differentiable point cloud rendering based on surface splatting. Our framework models the contribution of a point to the rendered image as a probability distribution. We derive an unbiased approximative gradient for the rendering function within this model. To efficiently evaluate the proposed sample estimate, we introduce a tree-based data-structure which employs multipole methods to draw samples in near linear time. Our gradient estimator allows us to avoid regularization required by previous methods, leading to a more faithful shape recovery from images. Furthermore, we validate that these improvements are applicable to real-world applications by refining the camera poses and point cloud obtained from a real-time SLAM system. Finally, employing our framework in a neural rendering setting optimizes both the point cloud and network parameters, highlighting the framework’s ability to enhance data driven approaches.

Files

978_3_031_19827_4_17.pdf
(pdf | 4.43 Mb)
- Embargo expired in 01-07-2023
License info not available