Poster De Conférence Année : 2026

J-NeuS: Joint field optimization for Neural Surface reconstruction in urban scenes with limited image overlap

Résumé

Reconstructing the surrounding surface geometry from recorded driving sequences poses a significant challenge due to the limited image overlap and complex topology of urban environments. SoTA neural implicit surface reconstruction methods often struggle in such setting, either failing due to small vision overlap or exhibiting suboptimal performance in accurately reconstructing both the surface and fine structures. To address these limitations, we introduce J-NeuS, a novel hybrid implicit surface reconstruction method for large driving sequences with outward facing camera poses. J-NeuS leverages cross-representation uncertainty estimation to tackle ambiguous geometry caused by limited observations. Our method performs joint optimization of two radiance fields in addition to guided sampling achieving accurate reconstruction of large areas along with fine structures in complex urban scenarios. Extensive evaluation on major driving datasets demonstrates the superiority of our approach in reconstructing large driving sequences with limited image overlap, outperforming concurrent SoTA methods.

Fichier non déposé

Dates et versions

hal-05499155 , version 1 (08-02-2026)

Identifiants

  • HAL Id : hal-05499155 , version 1

Citer

Fusang Wang, Hala Djeghim, Fabien Moutarde, Désiré Sidibé. J-NeuS: Joint field optimization for Neural Surface reconstruction in urban scenes with limited image overlap. Thirteenth International Conference on 3D Vision (3DV 2026), Mar 2026, Vancouver (BC), Canada. 2026. ⟨hal-05499155⟩
40 Consultations
0 Téléchargements

Partager

  • More