Newest papers

Neural Implicit Energy Optimization for LiDAR Mapping

Hrishikesh Viswanath, Md Ashiqur Rahman, Chi Lin, Damon Conover, Aniket Bera · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28598 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

As representações implícitas neurais fornecem parâmetros de campo de distância assinada (SDF) com eficiência de memória para cenas LiDAR de grande escala, mas a geometria recuperada de entradas escassas e muitas vezes ruidosas carrega jitter e artefatos. Técnicas clássicas de otimização geométrica resolvem tais artefatos com funções de energia de superfície, cujos minimizadores correspondem a superfícies lisas. Motivados por este princípio, formulamos o mapeamento neural LiDAR como a otimização de funções de energia sobre o espaço de parâmetro de representações implícitas neurais. Especificamente, o Weize regula o campo de distância assinado com uma função de energia definida no operador de forma de seu conjunto de nível, que representa a curvat

Machine translation (Meta M2M100); the English original follows.

Abstract

Neural implicit representations provide memory-efficient signed distance field (SDF) parameterizations for large-scale LiDAR scenes, but the geometry recovered from sparse and often noisy inputs carries jitter and artifacts. Classical geometric optimization techniques resolve such artifacts with surface energy functionals, whose minimizers correspond to smooth surfaces. Motivated by this principle, we formulate neural LiDAR mapping as the optimization of energy functionals over the parameter space of neural implicit representations. Specifically, we regularize the signed distance field with an energy functional defined on the shape operator of its level set, which represents the curvature of the surface, thus encouraging smoothness along the surface. This adapts classical geometric priors to the scale of million-point LiDAR maps. Experiments on large-scale datasets show reconstructions with improved smoothness at preserved geometric fidelity, and no additional inference-time cost. The code is publicly available at \href{https://github.com/HrishikeshVish/nieo}{https://github.com/HrishikeshVish/nieo}

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Lidar, Smoothness, Energy (signal processing), Curvature, Signed distance function, Artificial neural network

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange