MARIO: Motion-Augmented Real-Time Multi-Sensor Inertial Odometry

2026-06-01

AI Summary

This work addresses the susceptibility of pure inertial odometry to drift and noise during everyday human motion, as well as its lack of explicit modeling of motion dynamics. The authors propose a novel paradigm that integrates learned human pose priors with lightweight multimodal sensors—including magnetometers, barometers, and auxiliary IMUs—into an inertial odometry framework. By embedding physically consistent kinematic constraints and employing real-time optimization, the method enhances localization robustness. It represents the first approach to deeply fuse data-driven pose priors with multi-source lightweight sensing, substantially improving generalization. Evaluated on the Nymeria dataset, the proposed method reduces positional drift by up to 36%, and further decreases it to 42% when combined with multisensor fusion, significantly outperforming existing solutions.

Abstract

Inertial odometry (IO) using only Inertial Measurement Units (IMUs) provides a lightweight solution for human motion tracking in augmented reality (AR) and wearable devices. Recent learning-based IO methods have improved the generalizability of inertial localization through large-scale pretraining on human motion datasets. However, these approaches remain prone to drift and noise because they do not explicitly capture human motion dynamics, especially on daily activity datasets such as Nymeria. In this work, we propose to ground inertial odometry in human kinematics through a learned IMU-inferred pose prior, which promotes physically consistent motion constraints. We integrate this pose prior into existing IO architectures and reduce positional drift by up to 36% on the challenging Nymeria dataset, which is 5x larger than datasets used in prior work. We further improve long-term performance with a sensor-fusion framework that incorporates auxiliary signals from lightweight sensors already available on commercial AR glasses, including magnetometers, barometers, and secondary IMUs. With this fusion strategy, positional drift is reduced by up to 42%, improving robustness and generalization across diverse motion conditions. Together, our results introduce a new paradigm for inertial and lightweight odometry by unifying human motion kinematics with multimodal sensing, setting a new benchmark for accurate and robust camera-less human tracking. Our website is available at https://spice-lab.org/projects/MARIO/.

Problem

• inertial odometry
• motion drift
• human motion dynamics
• IMU
• sensor fusion

Innovation

• inertial odometry
• pose prior
• human kinematics
• sensor fusion
• motion tracking