# Inertial-Only Optimization for Visual-Inertial Initialization

arXiv: https://arxiv.org/abs/2003.05766 · HTML: https://mlsanity.com/paper/2003.05766/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Mar 2020.

## Built on by (most cited first, 11 of 11)

- [DM-VIO: Delayed Marginalization Visual-Inertial Odometry](https://mlsanity.com/paper/2201.04114/) (2022)
- [Open-vocabulary Mobile Manipulation in Unseen Dynamic Environments with 3D Semantic Maps](https://mlsanity.com/paper/2406.18115/) (2024)
- [HybVIO: Pushing the Limits of Real-time Visual-inertial Odometry](https://mlsanity.com/paper/2106.11857/) (2021)
- [Learned Monocular Depth Priors in Visual-Inertial Initialization](https://mlsanity.com/paper/2204.09171/) (2022)
- [EDI: ESKF-based Disjoint Initialization for Visual-Inertial SLAM Systems](https://mlsanity.com/paper/2308.02670/) (2023)
- [Trust Your IMU: Consequences of Ignoring the IMU Drift](https://mlsanity.com/paper/2103.08286/) (2021)
- [Improving Monocular Visual-Inertial Initialization with Structureless Visual-Inertial Bundle Adjustment](https://mlsanity.com/paper/2502.16598/) (2025)
- [GNSS-Inertial State Initialization Using Inter-Epoch Baseline Residuals](https://mlsanity.com/paper/2506.11534/) (2025)
- [ATDN vSLAM: An all-through Deep Learning-Based Solution for Visual Simultaneous Localization and Mapping](https://mlsanity.com/paper/2206.05963/) (2022)
- [Robust Structureless Monocular Visual Inertial Initialization Exploiting Line Features and Vanishing Points](https://mlsanity.com/paper/2609.21186/) (2026)
- [DAVIO: Dense Monocular-Inertial SLAM with Feed-Forward Initialization and Pose-Conditioned Mapping](https://mlsanity.com/paper/2609.27702/) (2026)

## Abstract (arXiv)

We formulate for the first time visual-inertial initialization as an optimal estimation problem, in the sense of maximum-a-posteriori (MAP) estimation. This allows us to properly take into account IMU measurement uncertainty, which was neglected in previous methods that either solved sets of algebraic equations, or minimized ad-hoc cost functions using least squares. Our exhaustive initialization tests on EuRoC dataset show that our proposal largely outperforms the best methods in the literature, being able to initialize in less than 4 seconds in almost any point of the trajectory, with a scale error of 5.3% on average. This initialization has been integrated into ORB-SLAM Visual-Inertial boosting its robustness and efficiency while maintaining its excellent accuracy.