Event-based Stereo Visual-Inertial Odometry with Voxel Map

TL;DR

This paper introduces Voxel-ESVIO, a stereo visual-inertial odometry system that uses voxel map management to select high-quality, noise-resilient 3D points, improving accuracy and efficiency in event-based odometry.

Contribution

The paper proposes a novel voxel-based point selection and management approach for event-based stereo visual-inertial odometry, enhancing map point quality and state estimation.

Findings

Outperforms state-of-the-art methods in accuracy.

Achieves higher computational efficiency.

Demonstrates robustness on public benchmarks.

Abstract

The event camera, renowned for its high dynamic range and exceptional temporal resolution, is recognized as an important sensor for visual odometry. However, the inherent noise in event streams complicates the selection of high-quality map points, which critically determine the precision of state estimation. To address this challenge, we propose Voxel-ESVIO, an event-based stereo visual-inertial odometry system that utilizes voxel map management, which efficiently filter out high-quality 3D points. Specifically, our methodology utilizes voxel-based point selection and voxel-aware point management to collectively optimize the selection and updating of map points on a per-voxel basis. These synergistic strategies enable the efficient retrieval of noise-resilient map points with the highest observation likelihood in current frames, thereby ensureing the state estimation accuracy. Extensive evaluations on three public benchmarks demonstrate that our Voxel-ESVIO outperforms state-of-the-art methods in both accuracy and computational efficiency.