# Benchmarking Visual Feature Representations for LiDAR-Inertial-Visual Odometry Under Challenging Conditions Source: [Benchmarking Visual Feature Representations for LiDAR-Inertial-Visual Odometry Under Challenging Conditions]() Collection index: [llms.txt]() Data updated: 2026-03-21T05:00:04.296Z Authors: Eunseon Choi, Junwoo Hong, Daehan Lee, Sanghyun Park, Hyunyoung Jo, Sunyoung Kim, Changho Kang, Seongsam Kim, Yonghan Jung, Jungwook Park, Seul Koo, Soohee Han Published: 2026-03-19T07:56:49.000Z ## 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. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Visual Odometry Translation Error Estimation on Newer College Quad-Hard](): Visual Odometry Translation Error Estimation; Newer College Quad-Hard; RMSE (m): 0.052. - [Visual Odometry Translation Error Estimation on Newer College Maths-Hard](): Visual Odometry Translation Error Estimation; Newer College Maths-Hard; RMSE (m): 0.061. - [Visual Odometry Translation Error Estimation on Super-Loc (Cave1)](): Visual Odometry Translation Error Estimation; Super-Loc (Cave1); RMSE (m): 0.185. - [Visual Odometry Translation Error Estimation on Super-Loc Cave2](): Visual Odometry Translation Error Estimation; Super-Loc Cave2; RMSE (m): 0.488. - [Absolute Translation Error on MARS-LVIG](): Absolute Translation Error; MARS-LVIG; ATE (AMtown01): not recorded. ## Original sources and implementation [Original paper]() [PDF]()