Distributionally Robust Surface Reconstruction from Sparse Point Clouds

Amine Ouasfi, Shubhendu Jena, Eric Marchand, Adnane Boukhayma

Keywords: implicit neural representations, 3D reconstruction from unoriented point could, Distributionally Robust Optimization

TL;DR: A distributionally Robust Optimization framework for SDF inference from sparse point-clouds

Abstract: We consider the problem of learning Signed Distance Functions (SDF) from sparse and noisy 3D point clouds. This task is significantly challenging when no ground-truth SDF supervision is available. Unlike recent approaches that rely on smoothness priors, our method, rooted in a distributionally robust optimization (DRO) framework, incorporates a regularization term that leverages samples from the uncertainty regions of the model to improve the learned SDFs. Thanks to tractable dual formulations, we show that this framework enables a stable and efficient optimization of SDFs in the absence of ground truth supervision. Through extensive experiments and evaluations, we illustrate the efficacy of our DRO inspired learning framework, highlighting its capacity to improve SDF learning with respect to baselines and the state-of-the-art using synthetic and real data evaluation.

Primary Area: applications to computer vision, audio, language, and other modalities

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 12865