UpDown-SC: Gravity-Canonicalized Dual-Envelope Scan Context for Indoor LiDAR Place Recognition

2026-09-24

AI Summary

This study addresses retrieval failures in indoor LiDAR localization caused by pose variations and ceiling interference. We propose a training-free polar scene descriptor that introduces gravity normalization to eliminate height discrepancies. For the first time, physical segmentation is leveraged to construct a dual-envelope complementary structure, jointly encoding upper- and lower-layer geometric features to enhance discriminability. Furthermore, a mask-aware non-uniform distance metric is designed to preserve ground-level evidence while suppressing cross-session variations. Experimental results demonstrate that the proposed method significantly improves retrieval reliability and F1 scores across diverse indoor scenarios. Notably, it supports lightweight real-time deployment on CPUs and facilitates map-based localization initialization.

Abstract

LiDAR place recognition is a key front end for loop closure and global relocalization, yet indoor retrieval remains difficult when attitude or sensor mounting height changes between mapping and query sessions. Scan Context stores the maximum height in each polar cell; indoors, broad ceilings can suppress the lower and mid-level geometry that distinguishes adjacent rooms and corridors. We present UpDown-SC, a training-free polar descriptor that first canonicalizes gravity and then represents two complementary surfaces: the upper envelope of lower/middle structures and the lower envelope of overhead structures. Their physical split is estimated once from a cell-balanced map height distribution and reused by every query. A mask-aware, non-uniform two-channel distance retains discriminative lower-level evidence while limiting sensitivity to its cross-session variation, without treating unobserved cells as zero-height measurements. Conventional Scan Context shortlisting and circular yaw alignment are retained, so retrieved hypotheses directly initialize geometric verification. Experiments across repeated indoor sessions, mounting-height changes, mixed outdoor-to-indoor trajectories, and an outdoor transfer sequence show more reliable first-choice retrieval on the indoor and mounting-height-varied sessions. A paired test finds a significant gain over Scan Context on the in-house sessions. UpDown-SC also gives the best or second-best F1max and AUPR under threshold-based acceptance while retaining a lightweight CPU front end. Continuous replay confirms that the retrieved hypotheses support metric prior-map localization. Code and evaluation artifacts: https://github.com/jiejie567/updown-sc.

Problem

• Indoor LiDAR place recognition
• Scan Context
• sensor mounting height variation
• attitude change
• loop closure

Innovation

• Place Recognition
• Scan Context
• Gravity Canonicalization
• Dual-Envelope Descriptor
• Indoor LiDAR