Benchmarking Visual Feature Representations for LiDAR-Inertial-Visual Odometry Under Challenging Conditions

2026-03-19

AI Summary

This work proposes a hybrid LiDAR-inertial-visual odometry approach that systematically integrates direct photometric methods with diverse learned and traditional feature descriptors—including ORB, SuperPoint, and XFeat—alongside their corresponding matching strategies such as Hamming distance, SuperGlue, and LightGlue, building upon the FAST-LIVO2 framework. By synergistically combining direct and feature-based paradigms, the method effectively mitigates the failure of pure direct methods in regions with inconsistent illumination, while maintaining low computational overhead. The resulting system significantly enhances feature tracking stability and localization accuracy, demonstrating robust and reliable state estimation even under challenging conditions such as low light, overexposure, varying illumination, and high scene disparity.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: State Estimation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate localization in autonomous driving is critical for successful missions including environmental mapping and survivor searches. In visually challenging environments, including low-light conditions, overexposure, illumination changes, and high parallax, the performance of conventional visual odometry methods significantly degrade undermining robust robotic navigation. Researchers have recently proposed LiDAR-inertial-visual odometry (LIVO) frameworks, that integrate LiDAR, IMU, and camera sensors, to address these challenges. This paper extends the FAST-LIVO2-based framework by introducing a hybrid approach that integrates direct photometric methods with descriptor-based feature matching. For the descriptor-based feature matching, this work proposes pairs of ORB with the Hamming distance, SuperPoint with SuperGlue, SuperPoint with LightGlue, and XFeat with the mutual nearest neighbor. The proposed configurations are benchmarked by accuracy, computational cost, and feature tracking stability, enabling a quantitative comparison of the adaptability and applicability of visual descriptors. The experimental results reveal that the proposed hybrid approach outperforms the conventional sparse-direct method. Although the sparse-direct method often fails to converge in regions where photometric inconsistency arises due to illumination changes, the proposed approach still maintains robust performance under the same conditions. Furthermore, the hybrid approach with learning-based descriptors enables robust and reliable visual state estimation across challenging environments.

Problem

• visual odometry
• challenging environments
• illumination changes
• LiDAR-inertial-visual odometry
• feature representation

Innovation

• hybrid visual odometry
• feature descriptor benchmarking
• LiDAR-inertial-visual odometry
• learning-based descriptors
• photometric consistency