F-LIVO: A Feature-Level Geometric Measurement Framework for LiDAR-Inertial-Visual Odometry

Xinye Dai; Yapeng Wang; Shujie Zhou; Zihao Wang; Shiqi Zheng; Jiayu Hu; Yusheng Wang; Weiwei Song

cs.ROOct 16, 2026

F-LIVO: A Feature-Level Geometric Measurement Framework for LiDAR-Inertial-Visual Odometry

Authors: Xinye Dai, Yapeng Wang, Shujie Zhou, Zihao Wang, Shiqi Zheng, Jiayu Hu, Yusheng Wang, Weiwei Song,

Organizations: The authors are with the GNSS Center, Wuhan University, Wuhan 430079, China.

Abstract

Measurement-level LiDAR-Inertial-Visual Odometry (LIVO) can achieve high accuracy, and its practical resource requirements depend on the adopted association and map-maintenance strategy. Persistently attaching dense photometric patches or appearance attributes to mapped LiDAR points can introduce additional storage, association, and update costs. This paper presents F-LIVO, a filter-based LIVO framework that couples a visual-inertial subsystem and a LiDAR-inertial subsystem through feature-level point-to-plane constraints between visual points and LiDAR planes. To preserve cross-modal coupling when visual triangulation is weak, we further recover visual feature depth from the LiDAR map through local plane fitting and ray-plane intersection. In the LiDAR update, the covariance of each point-to-plane residual is adaptively scaled according to local planar reliability, thereby reducing overconfident updates on poorly conditioned surfaces. Because the proposed method relies on sparse geometric features rather than persistent dense photometric map attributes, it retains a compact map representation. Experiments on multiple public datasets show competitive, sequence-dependent accuracy. Across the evaluated degradation cases, F-LIVO maintains continuous estimation with comparatively low accumulated end-to-end drift. Under the evaluated resource configuration, it achieves the lowest mean and peak central processing unit (CPU) usage, the lowest peak resident set size (RSS), and the lowest RSS growth rate among the evaluated implementations, while R2LIVE retains a lower mean RSS. The source code will be made publicly available upon acceptance to facilitate reproducible research.

Figures & tables

Notations	Explanation
$\boxplus/\boxminus$	The $\boxplus$ and $\boxminus$ operators on the state manifold
$(\cdot)^{G}$	A vector $(\cdot)$ expressed in the global frame
$(\cdot)^{C}$	A vector $(\cdot)$ expressed in the camera frame
$(\cdot)^{I}$	A vector $(\cdot)$ expressed in the IMU frame
$(\cdot)^{L}$	A vector $(\cdot)$ expressed in the LiDAR frame
$\mathbf{T}_{L}^{I}$	Rigid extrinsic transformation from the LiDAR frame to the IMU frame

TABLE I: Important Notations

Fig. 1: Overview of the proposed system. The system integrates LiDAR, IMU, and camera measurements within a unified filtering framework for LIVO. LiDAR and visual observations are coupled through point-cloud geometric information to achieve feature-level fusion. The resulting updates are applied to the relevant states in the filter, and the final output is the fused odometry estimate.

Panel of Figure 3 — Fig. 2: Illustrations of the visual-plane construction and LiDAR-assisted depth recovery. (a) Triangulated visual feature points. (b) Delaunay triangulation built from the tracked visual features. (c) Coplanar aggregation of neighboring triangles into visual feature planes. (d) LiDAR points projected onto the image plane are used for local search and plane fitting to recover feature depth when visual triangulation is weak. (e) Red points denote visual feature points whose depths are recovered from LiDAR geometry.

Fig. 3: Geometric illustration of the cross-modal coplanar association used in the camera update. Orange points denote LiDAR points and purple points denote visual feature points. When the two sets are associated with the same local plane, point-to-LiDAR-plane and LiDAR-point-to-visual-plane constraints can be constructed consistently.

Panel of Figure 5 — Fig. 4: Illustrations of the adaptive point-to-plane covariance and the unified residual construction. (a) Ideal point-to-plane distance constraint, where the point is assumed to lie on the plane. (b) Practical point-to-plane constraint, where the fitted plane has thickness and the point is generally not exactly on the plane. (c) Unified residual model used for feature-level LiDAR-visual fusion, where the blue and purple rays denote points associated with different plane features.

Fig. 5: (a) is the construction environment, rich in planar structure. (b) shows the campus environment, filled with piles of concrete and gravel. Both (a) and (b) have overexposure. (c) and (d) are basement environments with narrow passages and poor lighting conditions. (e) and (f) correspond to long-sequence scenarios.

	R3LIVE	R2LIVE	FAST-LIVO2	FAST-LIO2	ov_plane	F-LIVO (w/)	F-LIVO (w/o)
Basement1	17.5/31.2	19.9/29.5	15.8 /26.6	15.8/27.0	–/–	15.8 /28.0	15.9/28.3
Basement4	73.6/181.3	10.0/20.9	9.40/17.8	10.6/25.3	50.5/136.0	6.04 /11.5	57.2/126.1
Campus2	6.2/10.7	5.2/10.2	4.2 /10.1	7.0/16.8	62.4/90.1	5.7/9.7	8.0/17.7
Construction2	9.7/ 19.8	10.0/20.9	9.8/ 13.9	12.4/26.5	261.3/1054.1	8.2 /25.1	20.2/28.4
LabSurvey2	–/–	–/–	–/–	–/–	–/–	5.5 /25.8	–/–
UzhArea2	22.8 /38.7	26.4/43.5	23.6/39.7	29.8/48.5	–/–	23.5/42.4	26.8/46.0

TABLE II: The RMSE/MAX of ATE (UNIT: cm) results in HILTI and i2Nav-Robot dataset. The best results are marked in bold.

Fig. 6: Representative trajectory estimates on Basement4 for the compared methods. The repeated-evaluation ATE RMSE/MAX is reported in Table II .

Fig. 7: Representative trajectory estimates for F-LIVO and FAST-LIVO2 on Basement4 . The repeated-evaluation ATE RMSE/MAX is reported in Table II .

Fig. 8: Evolution of the planar quality factor $\sigma_{\mathrm{plane}}$ on the Basement4 sequence.

Fig. 9: F-LIVO’s trajectory in Building .

Fig. 10: Results on degenerate sequences 01 and 02 show that F-LIVO maintains continuous estimation with comparatively low accumulated end-to-end drift under the evaluated concurrent visual and LiDAR degradation.

Fig. 11: (a) R3LIVE partially recovers structure with slight misalignment. (b) R2LIVE exhibits pronounced distortion. (c) F-LIVO preserves the sharp-corner structure. (d) FAST-LIVO2 fails to follow the corner turn when the LiDAR faces the blank wall, resulting in an incomplete map.

Sequence	FAST-LIVO2	R3LIVE	R2LIVE	F-LIVO (w/o AC)	F-LIVO
Degenerate seq00	570.0/20.06	5.72/0.38	23.42/1.43	7.113/1.695	7.52/0.39
Degenerate seq01	1300.6/83.64	3.54/0.19	–/–	9.768/1.012	4.66/0.25
Degenerate seq02	–/–	8.46/0.28	–/–	21.947/2.251	13.68/0.23
LiDAR Degenerate	–/–	–/–	–/–	18.7/1.32	12.63/0.82
Visual Challenge	–/–	–/–	–/–	7.9/0.41	6.20/0.26

TABLE III: Accumulated end-to-end translation/rotation drift without loop closure (cm/deg). F-LIVO (w/o AC) denotes the fixed-noise variant without adaptive covariance weighting.

Fig. 12: Qualitative continuity evaluation in a tunnel-like degenerate environment. The red box marks the long, narrow, and repetitive tunnel segment, and the enlarged view shows the corresponding local reconstruction. The pose output and reconstruction remain continuous through the highlighted segment. Since no complete ground-truth trajectory is available, the figure provides qualitative evidence of estimator continuity rather than trajectory-wide accuracy.

	FAST-LIVO2	F-LIVO	R3LIVE
Drift (m/deg)	2.01 / 0.77	2.03 / 0.74	2.19 / 1.89

TABLE IV: Odometry drift in large-scale environment.

Panel of Figure 17 — Fig. 13: Process-level resource usage on hku_campus_seq_00 under the unified profiling protocol. (a) Aggregate process CPU usage, where 100% corresponds to one logical core. (b) Resident set size (RSS). All methods are evaluated over the same 170-s interval after excluding the first 10 s for initialization.

Method	Mean CPU (%)	Peak CPU (%)	Mean RSS (GB)	Peak RSS (GB)	Growth rate (MB/min)
F-LIVO	104.99	137.65	1.250	1.322	36.06
FAST-LIVO2	243.80	332.87	3.052	5.679	1909.48
R3LIVE	331.52	425.49	2.611	4.742	1501.62
R2LIVE	206.35	248.57	0.976	1.451	318.49

TABLE VI: Process-level resource statistics over the common 170-s analysis interval. CPU is aggregate process CPU, where 100% corresponds to one logical core; memory is process RSS.

State Size	15	30	60	120
R3LIVE	0.034	0.067	0.209	0.998
FAST-LIVO2	0.023	0.024	0.026	0.028
F-LIVO	0.024	0.024	0.025	0.026

TABLE V: MEASUREMENT-UPDATE TIME UNDER DIFFERENT STATE DIMENSIONS (MS).

Camera-LiDAR Fusion

Camera–LiDAR fusion combines image semantics with LiDAR geometry for 3D perception, depth estimation, localization, and mapping.

See all 103 papers

SA-LIVO: Efficient LiDAR-Inertial-Visual Odometry with Subspace-Aware Degeneracy Handling
Jun 24, 2026Yinong Cao, Xin He, Yijie Wang +11Inertial Odometry Robot Localization
Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China. · Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China.
PA-LVIO: Real-Time LiDAR-Visual-Inertial Odometry and Mapping with Pose-Only Bundle Adjustment
Mar 17, 2026Hailiang Tang, Tisheng Zhang, Liqiang Wang +3Adaptive Lidar-Inertial Odometry Simultaneous Localization and Mapping
the GNSS Research Center, Wuhan University, Wuhan 430079, China · the Hubei Technology Innovation Center for Spatiotemporal Information and Positioning Navigation, Wuhan 430079, China · the Hubei Luojia Laboratory, Wuhan 430079, China
FAST-LIVGO: A Degeneracy-Robust LiDAR-Inertial-Visual-GNSS Fusion Odometry
Jun 17, 2026Zhiyu Chen, Chunran Zheng, Jiayu Wen +4Error-State Kalman Filtering Robot Localization
College of Mechatronics and Control Engineering, Shenzhen University, Shenzhen 518060, China · Department of Mechanical Engineering, The University of Hong Kong, Hong Kong SAR, China · College of Automation, Harbin Engineering University, Harbin 150001, China

Jun 24, 2026J/K move · Enter open · S save

SA-LIVO: Efficient LiDAR-Inertial-Visual Odometry with Subspace-Aware Degeneracy Handling

Yinong Cao, Xin He, Yijie Wang, Tianzhen Ma+10

Key Laboratory of Space Active Opto-Electronics Technology, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China. · Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China.

Inertial Odometry Robot Localization Adaptive Lidar-Inertial Odometry Multimodal Sensor Fusion

Open paper AskPDF arXiv

Light detection and ranging (LiDAR)-inertial-visual odometry fuses geometric depth with visual measurements, but its sensors fail independently: LiDAR in under-constrained geometry, vision under poor illumination or texture. Weak information is harmless, since the filter then relies on the other sensor or the inertial prediction; over-confident information is not. Existing countermeasures act on weak information or switch vision by hand-set thresholds. This article presents SA-LIVO, which fuses vision through explicit, online-estimated error models. A frame-level noise scale, estimated by covariance matching, absorbs whole-image errors, and Schmidt filtering carries the correlation of reused observations. Subspace-aware information fusion models the error shared by a frame's observations as a pose offset capping the visual information per direction; the exact fusion attenuates vision only above that cap and reduces to the information sum when the visual model is consistent. LiDAR and visual residuals share one invariant extended Kalman filter loop. On 24 handheld and aerial sequences, SA-LIVO is the most accurate or tied on 22 and stays bounded where others diverge; on the HILTI 2022 benchmark it needs 13.4 ms per frame on a laptop processor and 2.0-6.2x less memory than FAST-LIVO2, R3LIVE, and SR-LIVO.

Abstract

Camera-LiDAR Fusion

Explore similar work

SA-LIVO: Efficient LiDAR-Inertial-Visual Odometry with Subspace-Aware Degeneracy Handling

PA-LVIO: Real-Time LiDAR-Visual-Inertial Odometry and Mapping with Pose-Only Bundle Adjustment

FAST-LIVGO: A Degeneracy-Robust LiDAR-Inertial-Visual-GNSS Fusion Odometry