Visibility-Aware Reference Generation for Benchmarking Visual 3D Reconstruction in Vegetation-Rich Environments
Visibility-Aware Reference Generation for Benchmarking Visual 3D Reconstruction in Vegetation-Rich Environments
This paper presents a LiDAR-based pipeline to generate reference depth maps and point clouds from existing multi-sensor localization data, introducing the LargeDepth-Gap Visibility Score (LDGVS) for visibility filtering. Experiments on TartanAirV2, TartanGround, and BotanicGarden show that combining Hidden Point Removal with LDGVS retains approximately three times the point density of Poisson surface reconstruction at comparable error levels, while comparing feed-forward depth estimation against conventional SfM methods.
The authors introduce a LiDAR-based reference-generation pipeline and the LargeDepth-Gap Visibility Score (LDGVS) to benchmark visual 3D reconstruction in vegetation-rich environments, comparing feed-forward methods with conventional SfM on TartanAirV2, TartanGround, and BotanicGarden datasets.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2611.05361v1 Announce Type: cross Abstract: Reliable evaluation of visual 3D reconstruction and depth estimation in vegetation-rich outdoor environments is limited by the scarcity of geometric ground-truth data and the cost of complete scanner surveys. We present a LiDARbased reference-generation pipeline for evaluating the zero-shot performance of feed-forward visual-geometry models alongside conventional stru
Open source S2
cture-from-motion (SfM) methods. The pipeline reuses camera images, LiDAR scans, and reference poses from existing multi-sensor localization datasets to generate per-view depth maps and visible reference point clouds. To address visibility errors in accumulated LiDAR scans, we introduce the LargeDepth-Gap Visibility Score (LDGVS) for point-wise visibility filtering and automatic scale selection in Hidden Point Remova
Open source S3
l(HPR). Experiments on TartanAirV2 and TartanGround show that, compared with Poisson surface reconstruction, which is an approach suited to completed scan surveys, the combined HPR and LDGVS filtering retains approximately three times the point density at comparable absolute relative error. We also apply the pipeline to the BotanicGarden dataset to systematically compare depth estimation and 3D reconstruction methods
Open source S4
. Feedforward methods produce dense depth predictions efficiently, but their 3D reconstructions can contain duplicated structures. Conventional SfM generally produces more coherent geometry, but its reconstructions remain sparse and more prone to failure in challenging vegetated scenes. The benchmark toolkit and datasets will be released.
Open source S5
Visibility-Aware Reference Generation for Benchmarking Visual 3D Reconstruction in Vegetation-Rich Environments
Open source S1
Source：arXiv Robotics — research abstracts · arxiv.org