POPL-KF: A Pose-Only Geometric Representation-Based Kalman Filter for Point-Line-Based Visual-Inertial Odometry

TL;DR

POPL-KF tackles the limitations of traditional VIO by introducing pose-only representations for both point and line features, removing feature coordinates from measurement equations to reduce linearization errors and enable immediate updates. It integrates a unified base-frame selection strategy and a grid-plus-bidirectional-flow line filter, achieving robust, real-time performance on diverse datasets and real-world scenarios. The approach outperforms state-of-the-art filter- and optimization-based methods in localization accuracy and robustness, particularly in low-texture or challenging environments where line geometry is plentiful. This work advances practical visual-inertial odometry for resource-constrained platforms by leveraging pose-only measurements and improved line-feature handling.

Abstract

Mainstream Visual-inertial odometry (VIO) systems rely on point features for motion estimation and localization. However, their performance degrades in challenging scenarios. Moreover, the localization accuracy of multi-state constraint Kalman filter (MSCKF)-based VIO systems suffers from linearization errors associated with feature 3D coordinates and delayed measurement updates. To improve the performance of VIO in challenging scenes, we first propose a pose-only geometric representation for line features. Building on this, we develop POPL-KF, a Kalman filter-based VIO system that employs a pose-only geometric representation for both point and line features. POPL-KF mitigates linearization errors by explicitly eliminating both point and line feature coordinates from the measurement equations, while enabling immediate update of visual measurements. We also design a unified base-frames selection algorithm for both point and line features to ensure optimal constraints on camera poses within the pose-only measurement model. To further improve line feature quality, a line feature filter based on image grid segmentation and bidirectional optical flow consistency is proposed. Our system is evaluated on public datasets and real-world experiments, demonstrating that POPL-KF outperforms the state-of-the-art (SOTA) filter-based methods (OpenVINS, PO-KF) and optimization-based methods (PL-VINS, EPLF-VINS), while maintaining real-time performance.

Figures (12)

• Figure 1: System overview of POPL-KF. The light pink-filled boxes represent the processing steps of pose-only line features.
• Figure 2: Illustration of pose-only representation for line features.
• Figure 3: Illustration of immediate and delayed update.
• Figure 4: The images sampled from the three datasets.
• Figure 5: A mobile vehicle.