Vision-Aided Absolute Trajectory Estimation Using an Unsupervised Deep Network with Online Error Correction

TL;DR

This paper introduces VIOLearner, an unsupervised deep neural network that fuses RGB-D images with inertial data for accurate, calibration-free absolute trajectory estimation, with online error correction to enhance performance.

Contribution

The paper presents a novel unsupervised deep learning method for visual-inertial odometry that does not require IMU calibration or extrinsic camera-IMU calibration, and incorporates online error correction.

Findings

Achieves competitive odometry accuracy on KITTI dataset.

Does not require IMU intrinsic or extrinsic calibration.

Outperforms some existing visual-inertial odometry methods.

Abstract

We present an unsupervised deep neural network approach to the fusion of RGB-D imagery with inertial measurements for absolute trajectory estimation. Our network, dubbed the Visual-Inertial-Odometry Learner (VIOLearner), learns to perform visual-inertial odometry (VIO) without inertial measurement unit (IMU) intrinsic parameters (corresponding to gyroscope and accelerometer bias or white noise) or the extrinsic calibration between an IMU and camera. The network learns to integrate IMU measurements and generate hypothesis trajectories which are then corrected online according to the Jacobians of scaled image projection errors with respect to a spatial grid of pixel coordinates. We evaluate our network against state-of-the-art (SOA) visual-inertial odometry, visual odometry, and visual simultaneous localization and mapping (VSLAM) approaches on the KITTI Odometry dataset and demonstrate competitive odometry performance.