PulseAugur
LIVE 16:09:06
research · [2 sources] ·
0
research

Paired-CSLiDAR benchmark refines aerial-ground LiDAR pose with height-stratified registration

Researchers have developed Paired-CSLiDAR, a new benchmark and registration method for refining the pose of ground-based LiDAR scans using aerial scans. The challenge lies in the limited shared geometry between aerial and ground perspectives, often only the terrain surface. Their proposed Residual-Guided Stratified Registration (RGSR) pipeline, which is training-free and geometry-only, leverages height-stratified ICP and other techniques to improve accuracy. RGSR demonstrated superior performance on the benchmark, outperforming existing methods like confidence-gated cascade and GeoTransformer. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Introduces a novel benchmark and method for improving LiDAR pose accuracy in challenging cross-source scenarios.

RANK_REASON This is a research paper detailing a new benchmark and registration method for LiDAR pose refinement.

Read on arXiv cs.CV →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 · Montana Hoover, Jing Liang, Tianrui Guan, Dinesh Manocha ·

    Paired-CSLiDAR: Height-Stratified Registration for Cross-Source Aerial-Ground LiDAR Pose Refinement

    arXiv:2605.00634v1 Announce Type: cross Abstract: We introduce Paired-CSLiDAR (CSLiDAR), a cross-source aerial-ground LiDAR benchmark for single-scan pose refinement: refining a ground-scan pose within a 50 m-radius aerial crop. The benchmark contains 12,683 ground-aerial pairs a…

  2. arXiv cs.CV TIER_1 · Dinesh Manocha ·

    Paired-CSLiDAR: Height-Stratified Registration for Cross-Source Aerial-Ground LiDAR Pose Refinement

    We introduce Paired-CSLiDAR (CSLiDAR), a cross-source aerial-ground LiDAR benchmark for single-scan pose refinement: refining a ground-scan pose within a 50 m-radius aerial crop. The benchmark contains 12,683 ground-aerial pairs across 6 evaluation sites and per-scan reference 6-…