This page contains a collection of papers that provide a collection of perspectives on a particular subject. Many might be outdated by now, but this was mainly for my own use rather then others. You should be able to find each paper by just searching the title of it.

Visual-Inertial State Initialization

  • Estimator initialization in vision-aided inertial navigation with unknown camera-IMU calibration - Link
  • Closed-form Solutions for Vision-aided Inertial Navigation - Link
  • Closed-form solution of visual-inertial structure from motion - Link
  • Simultaneous state initialization and gyroscope bias calibration in visual inertial aided navigation - Link
  • Robust initialization of monocular visual-inertial estimation on aerial robots - Link
  • A convex formulation for motion estimation using visual and inertial sensors - Link
  • VINS-Mono: A robust and versatile monocular visual-inertial state estimator - Link
  • Visual-inertial monocular SLAM with map reuse - Link
  • Fast and Robust Initialization for Visual-Inertial SLAM - Link
  • Inertial-only optimization for visual-inertial initialization - Link
  • Monocular visual–inertial state estimation with online initialization and camera–IMU extrinsic calibration - Link
  • Revisiting visual-inertial structure from motion for odometry and SLAM initialization - Link
  • An Analytical Solution to the IMU Initialization Problem for Visual-Inertial Systems - Link
  • Mid-Air Range-Visual-Inertial Estimator Initialization for Micro Air Vehicles - Link

Open-Sourced Visual-Inertial Codebases

  • rpng / OpenVINS - Link
  • rpng / R-VIO - Link
  • ethz-asl / okvis - Link
  • ethz-asl / maplab - Link
  • ethz-asl / rovio - Link
  • TUM / basalt - Link
  • HKUST-Aerial-Robotics / VINS-Fusion - Link
  • HKUST-Aerial-Robotics / VINS-Mono - Link
  • MIT-SPARK / Kimera-VIO - Link
  • ucla-vision / xivo - Link
  • KumarRobotics / msckf_vio - Link
  • jpl-x / x - Link

Continuous-Time Trajectory Estimation

  • Unified temporal and spatial calibration for multi-sensor systems - Link
  • Spline Fusion: A continuous-time representation for visual-inertial fusion with application to rolling shutter cameras. - Link
  • Continuous-time visual-inertial odometry for event cameras - Link
  • Efficient Visual-Inertial Navigation using a Rolling-Shutter Camera with Inaccurate Timestamps. - Link
  • Alternating-stereo VINS: Observability analysis and performance evaluation - Link
  • Multi-camera visual-inertial navigation with online intrinsic and extrinsic calibration - Link
  • Decoupled Representation of the Error and Trajectory Estimates for Efficient Pose Estimation - Link

Machine Learning Uncertainty

  • Uncertainty in Deep Learning - Link
  • Geometry and uncertainty in deep learning for computer vision - Link
  • Modelling uncertainty in deep learning for camera relocalization - Link
  • Dropout as a bayesian approximation: Representing model uncertainty in deep learning - Link
  • What uncertainties do we need in bayesian deep learning for computer vision? - Link
  • Multi-task learning using uncertainty to weigh losses for scene geometry and semantics - Link

Resource Constrained Extended Kalman Filtering

  • A provably consistent method for imposing sparsity in feature-based SLAM information filters - Link
  • Optimization-based estimator design for vision-aided inertial navigation - Link
  • Vision-aided inertial navigation for resource-constrained systems - Link
  • Power-SLAM: a linear-complexity, anytime algorithm for SLAM - Link
  • A resource-aware vision-aided inertial navigation system for wearable and portable computers - Link
  • An iterative kalman smoother for robust 3D localization and mapping - Link
  • Inverse Schmidt Estimators - Link
  • Consistent map-based 3D localization on mobile devices - Link
  • RISE-SLAM: A Resource-aware Inverse Schmidt Estimator for SLAM - Link
  • Markov Parallel Tracking and Mapping for Probabilistic SLAM - Link

Event-based Cameras

  • Event-based Vision Resources (master list) - Link
  • Real-Time 3D Reconstruction and 6-DoF Tracking with an Event Camera - Link
  • Continuous-Time Trajectory Estimation for Event-based Vision Sensors - Link
  • Continuous-Time Visual-Inertial Trajectory Estimation with Event Cameras - Link
  • EVO: A Geometric Approach to Event-Based 6-DOF Parallel Tracking and Mapping in Real-time - Link
  • Event-based Visual Inertial Odometry - Link
  • EEKLT: Asynchronous, Photometric Feature Tracking using Events and Frames - Link

Rolling Shutter Cameras

  • MIMC-VINS: A Versatile and Resilient Multi-IMU Multi-Camera Visual-Inertial Navigation System - Link
  • Rolling Shutter Camera Calibration - Link
  • Continuous-time batch trajectory estimation using temporal basis functions - Link
  • Vision-aided inertial navigation with rolling-shutter cameras - Link
  • Efficient Visual-Inertial Navigation using a Rolling-Shutter Camera with Inaccurate Timestamps - Link
  • Real-time Motion Tracking on a Cellphone using Inertial Sensing and a Rolling-Shutter Camera - Link
  • 3-D Motion Estimation and Online Temporal Calibration for Camera-IMU Systems - Link
  • High-fidelity Sensor Modeling and Self-Calibration in Vision-aided Inertial Navigation - Link

Historical VINS Literature

  • A Robust and Modular Multi-Sensor Fusion Approach Applied to MAV Navigation - Link
  • Determining the Time Delay Between Inertial and Visual Sensor Measurements - Link
  • A Multi-State Constraint Kalman Filter for Vision-aided Inertial Navigation - Link
  • Vision-Aided Inertial Navigation for Spacecraft Entry, Descent, and Landing - Link

Original ORB-SLAM

  • ORB: an efficient alternative to SIFT or SURF - Link
  • Bags of Binary Words for Fast Place Recognition in Image Sequences - Link
  • ORB-SLAM: Tracking and Mapping Recognizable Features - Link
  • ORB-SLAM: a Versatile and Accurate Monocular SLAM System - Link
  • Parallel Tracking and Mapping for Small AR Workspaces - Link

Go Home