SCILLA: SurfaCe Implicit Learning for Large Urban Area, a volumetric hybrid solution

arXiv:2403.10344v1 Announce Type: new
Abstract: Neural implicit surface representation methods have recently shown impressive 3D reconstruction results. However, existing solutions struggle to reconstruct urban outdoor scenes due to their large, unbounded, and highly detailed nature. Hence, to achieve accurate reconstructions, additional supervision data such as LiDAR, strong geometric priors, and long training times are required. To tackle such issues, we present SCILLA, a new hybrid implicit surface learning method to reconstruct large driving scenes from 2D images. SCILLA's hybrid architecture …

3d reconstruction abstract arxiv cs.cv data however hybrid lidar nature representation results solution solutions struggle supervision surface type urban