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

Jie Xu; Yongxin Yang; Ziyi Jin; Kangjin Yu; Hongjun Huang; Chao Han; Zhongpu Xia

cs.CVSep 24, 2026

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

Authors: Jie Xu, Yongxin Yang, Ziyi Jin, Kangjin Yu, Hongjun Huang, Chao Han, Zhongpu Xia

Organizations: Anyverse Dynamics, Beijing, China

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.

Figures & tables

Fig. 1: Why maximum-height SC loses indoor structure. Top: a gravity-canonicalized side view of one measured 0.1-s indoor query, colored by height. The dashed line marks the map split $\tau_{g}$ ; the display omits far and above-ceiling returns. Where lower and overhead structure share a polar cell, SC keeps only the upper maximum. UpDown-SC keeps both envelopes. Lower panels are schematics, not measured performance.

Fig. 2: End-to-end pipeline on one measured IH query from the 2 m evaluation. The shared map/query front end canonicalizes gravity and builds the dual-envelope SCD. Mask-aware retrieval returns ranked place and yaw hypotheses (here $\widehat{\Delta\psi}=180^{\circ}$ ), and point-to-map ICP verifies each 6-DoF seed. Panel (d) overlays the query on the retrieved keyframe using only the descriptor seed, before ICP. Retrieval metrics are measured before ICP.

Panel of Figure 3 — Fig. 3: One measured query shown at a common gravity-canonicalized reference. (a) White-background top view with a display-only crop of sparse range-boundary returns; all valid points still enter the descriptor, and the orange marker is its origin. (b) Conventional SC is dominated by overhead maxima in 431 of 467 occupied cells; 197 of these cells also contain lower/middle returns. (c) Up retains the overhead layer and Down separately retains the lower/middle layer. Descriptor heights share a color scale capped at 4.0 m, with the triangular tip denoting larger values.

Fig. 4: Rectangular SC-style feature maps of the same query as Fig. 3 , sharing one $16\times 60$ grid, color scale, and circular shift (Up drawn with near rings at the top, as in Fig. 2 ). Up retains SC’s overhead support (431 cells, 92.3%); Down restores all 197 lower/middle observations hidden by SC’s full-height maximum. Data-selected 12-sector windows mark three advantages: magenta, the densest mixed-cell region (59 cells); green dashed, 49 overhead cells whose stored maximum and visible underside differ by a median 2.2 m (kept only by Up); black dotted, lower structure spreading 0.90 m where overhead pins SC’s spread to 0.35 m. No value is interpolated or edited.

Parameter	Value
Descriptor construction
Rings / sectors	16 / 60
Maximum radius	30 m
Voxel size	0.25 m
Adaptive split range / bin	$[1.5,4.5]$ m / 0.1 m
Adaptive split: support / map keyframes	5% / 20

TABLE I: Core descriptor and retrieval settings.

	IH pilot ( $n_{q}=320$ )		CH public ( $n_{q}=148$ )		Indoor deployment (+G)		M2DGR (+G)	NC outdoor (+G)	Query latency ( $n_{q}=92$ )
Method	Native	+G	Native	+G	H1 $\rightarrow$ H2 ( $n_{q}=58$ )	H1 $\rightarrow$ V1 ( $n_{q}=57$ )	$n_{q}=28$	$n_{q}=60$	Median (ms) $\downarrow$
SC	59.1/74.1	62.5/79.4	$\mathbf{51.4}$ /66.9	60.1/82.4	39.7/67.2	47.4/78.9	$\mathbf{96.4}$ / 96.4	98.3 / $\mathbf{100.0}$	$\mathbf{3.6}$
SC++ (PC)	53.4/76.9	54.7/77.8	39.9/68.9	62.8/ 87.2	31.0/51.7	50.9 /80.7	75.0/ $\mathbf{100.0}$	96.7/ $\mathbf{100.0}$	8.8
SOLiD	35.6/57.5	34.4/55.6	35.8/ $\mathbf{75.0}$	42.6/69.6	39.7/60.3	36.8/61.4	92.9 / $\mathbf{100.0}$	53.3/88.3	13.3
M2DP †	47.8/63.7	50.0/64.1	48.6 /68.2	45.9/63.5	6.9/24.1	5.3/28.1	46.4/85.7	93.3/ $\mathbf{100.0}$	25.5
LiDAR Iris	58.1/72.2	61.9/76.2	42.6/67.6	$\mathbf{66.2}$ / $\mathbf{91.2}$	44.8 / 69.0	$\mathbf{61.4}$ / $\mathbf{86.0}$	85.7/92.9	95.0/ $\mathbf{100.0}$	515.2

TABLE II: Indoor and outdoor retrieval (R@1/R@5, %).

Panel of Figure 7 — Fig. 5: Top-1 retrievals for all 320 eligible IH queries (shared +G protocol): blue circles correct, red crosses failures linked to their selected keyframes, gray trajectories.

Fig. 6: Top-1 acceptance precision–recall under the shared gravity-canonicalized protocol. UpDown-SC attains the best F1max/AUPR on IH (0.696/0.475, vs. 0.643/0.398 for RING++) and the second-best on CH (0.644/0.530) behind LiDAR Iris (0.671/0.587) at roughly

40\times

its query latency. BTC lacks a comparable scalar confidence.

Panel of Figure 9 — Fig. 7: Channel-weight sensitivity over 11 pairs. (a) Macro Recall@1/5; dashed: IH+G-pilot selection $(w_{\ell},w_{h})=(0.3,0.7)$ . (b) Recall@1 change from equal weighting for the seven Table II conditions; circles mark row maxima and the outline marks the selected pair.

Explore similar work

InLiER: Learning-Free Heterogeneous LiDAR Place Recognition via Intermediate Mixed-Radix Structural Keypoint Tokenization
Jul 18, 2026Nikolaos Stathoulopoulos, George NikolakopoulosLidar Visual Place Recognition
Robotics and AI Group, Department of Computer, Electrical and Space Engineering, Luleå University of Technology, 971 87 Luleå, Sweden
Aerial-ground LiDAR place recognition with patch-level self-supervised learning and expanded reciprocal re-ranking
Jun 17, 2026Yandi Yang, Xianghong Zou, Jianping Li +4Lidar Visual Place Recognition
Department of Geomatics Engineering, University of Calgary, Calgary, T2N 1N4, Canada · School of Advanced Manufacturing, Nanchang University, Nanchang, 330031, China · School of Electrical and Electronic Engineering, Nanyang Technological University, 639798, Singapore +1
ReLoc: Rethinking Scene Coordinate Regression Architecture for Robust Outdoor LiDAR-based Localization
Sep 27, 2026Heejoon Moon, Yurim Cho, Je Hyeong HongScene Reconstruction Human Relocalization
Department of Artificial Intelligence, Hanyang University · Department of Electronic Engineering, Hanyang University

Jul 18, 2026, cs.CVJ/K move · Enter open · S save

InLiER: Learning-Free Heterogeneous LiDAR Place Recognition via Intermediate Mixed-Radix Structural Keypoint Tokenization

Nikolaos Stathoulopoulos, George Nikolakopoulos

Robotics and AI Group, Department of Computer, Electrical and Space Engineering, Luleå University of Technology, 971 87 Luleå, Sweden

Lidar Visual Place Recognition Outliers

Open paper AskPDF arXiv

LiDAR place recognition supports loop closure, relocalization, and multi-agent map management. As robotic platforms increasingly combine LiDARs with different fields of view, resolutions, and scanning patterns, existing descriptors degrade because they are tightly coupled to sensor-specific characteristics. We present InLiER, a learning-free pipeline based on an intermediate tokenization step. Height-sliced keypoints from structural elements receive mixed-radix token IDs encoding height, radial distance, local shape, and azimuth from local 3D geometry, in a compact sub-2KB representation. The same vocabulary is reorganized across three retrieval stages: height-ceiling histogram intersection for fast rotation-invariant shortlisting, binary bitmask alignment for yaw estimation and reranking, and token-guided geometric verification for 6-DoF pose estimation. InLiER achieves state-of-the-art performance on the HeLiPR dataset and in real-world field experiments, among modern handcrafted methods and outperforms the learning-based baseline on most cross-sensor configurations.

Abstract

Explore similar work

InLiER: Learning-Free Heterogeneous LiDAR Place Recognition via Intermediate Mixed-Radix Structural Keypoint Tokenization

Aerial-ground LiDAR place recognition with patch-level self-supervised learning and expanded reciprocal re-ranking

ReLoc: Rethinking Scene Coordinate Regression Architecture for Robust Outdoor LiDAR-based Localization