Toward Robust Neural Reconstruction from Sparse Point Sets

Autor:	Ouasfi, Amine, Jena, Shubhendu, Marchand, Eric, Boukhayma, Adnane
Rok vydání:	2024
Předmět:	Computer Science - Computer Vision and Pattern Recognition
Druh dokumentu:	Working Paper
Popis:	We consider the challenging problem of learning Signed Distance Functions (SDF) from sparse and noisy 3D point clouds. In contrast to recent methods that depend 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. Using a variety of synthetic and real data evaluations from different modalities, we show that our DRO based learning framework can improve SDF learning with respect to baselines and the state-of-the-art methods. Comment: Project page : https://ouasfi.github.io/sdro/
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2412.16361 Zobrazit plný text záznamu View this record from Arxiv