Event-Frame-Inertial Odometry Using Point and Line Features Based on Coarse-to-Fine Motion Compensation (C2F-EFIO)
Authors : Byeongpil Choi, Hanyeol Lee, and Chan Gook Park
This repository contains the ROS package for Event-Frame-Inertial Odometry Using Point and Line Features Based on Coarse-to-Fine Motion Compensation (C2F-EFIO), written in C++. [Paper]
- This package was tested on Ubuntu 16.04 (ROS Kinetic) with C++11.
- Dependencies:
- Eigen 3.3.7
- OpenCV 3.3.1
- Boost 1.58.0.1
- No additional dependencies are required, and we expect this package to build without issues in other environments.
- The package uses the catkin build system:
cd catkin_ws
catkin_make
- Expected directory structure:
catkin_ws
│ src
│ ├── config
│ ├── dvs_msgs
│ ├── mono_tracker
│ ├── msckf_estimator
│ └── rviz
│ build
│ devel
│ dataset
│ ├── ECDS
│ │ ├── boxes_6dof.bag
│ │ ├── ...
│ │ ├── shapes_translation.bag
│ └── droneracing
│ ├── indoor_45_2_davis_with_gt.bag
│ ├── ...
│ ├── indoor_forward_10_davis_with_gt.bag
- Prepare the visual-inertial calibration file in the
config/folder and set the appropriate ROS topics inmsckf_estimator/launch/.
-
Launch files:
msckf_estimator/smsckf_nesl_ECDS.launch: Event-Camera Datasetmsckf_estimator/smsckf_nesl_droneracing_IF.launch: UZH-FPV Drone Racing Indoor Forward Datasetmsckf_estimator/smsckf_nesl_droneracing_I45.launch: UZH-FPV Drone Racing Indoor 45-Degree Datasetmsckf_estimator/smsckf_nesl_rviz.launch: RViz Launch
Parameters setting in .launch files:
max_cnt: The maximum number of point features to be tracked max_events: The number of events used to create the initial event frame (adaptively adjusted during runtime) life_pixel: The number of pixels a feature is allowed to move between consecutive event frames to calculate the lifetime of event point features eta: A ratio factor used to determine the optimal window size for extracting event lines pix_th: A threshold for selecting informative event points ransac_iter: The maximum number of iterations for RANSAC-based line fitting dist_thr: The distance threshold used for line outlier rejection tests theta_thr: The angle threshold used for line outlier rejection tests
-
Launch the launch files
roslaunch smsckf_nesl smsckf_nesl_ECDS.launch --screen roslaunch smsckf_nesl smsckf_nesl_rviz.launch rosbag play /root/catkin_ws/dataset/ECDS/boxes_6dof.bag # prepared dataset -
Save the results
rosbag record /smsckf_nesl/odom : img-synced output
- Line detection
- Pose estimation
If you feel this work helpful to your academic research, we kindly ask you to cite our paper :
@article{C2F-EFIO_RAL,
title={Event-Frame-Inertial Odometry Using Point and Line Features Based on Coarse-to-Fine Motion Compensation},
author={Choi, Byeongpil and Lee, Hanyeol and Park, Chan Gook},
journal={IEEE Robotics and Automation Letters},
year={2025},
publisher={IEEE}
}
This research was supported by the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT, the Republic of Korea. (NRF-2022R1A2C2012166)
Our source code is released under MIT license. If you encounter any issues or have questions, please contact the author at bpc1224@snu.ac.kr.