VINS-Mask

A ROI-mask Feature Tracker for Monocular Visual-Inertial System, ICARCE 2022.

Project Page

1. Prerequisites

1.1 Ubuntu and ROS Ubuntu 16.04. ROS Kinetic. ROS Installation additional ROS pacakge

    sudo apt-get install ros-YOUR_DISTRO-cv-bridge ros-YOUR_DISTRO-tf ros-YOUR_DISTRO-message-filters ros-YOUR_DISTRO-image-transport

1.2 CUDA and pytorch Follow Pytorch Installation.

1.3 Ceres Solver Follow Ceres Installation, remember to sudo make install.

1.4 gRPC and Protocol Buffers Follow gRPC Installation, remember to sudo make install. (Our testing environment: Ubuntu 16.04, ROS Kinetic, OpenCV 3.3.1, Protobuf 3.11.2, gRPC 1.9.0, torch 1.8.2, CUDA 11.1)

2. Build VINS-Mask on ROS

Clone the repository and catkin_make:

    cd ~/catkin_ws/src
    git clone https://github.com/sunjiayuanro/VINS-Mask.git
    cd ../
    catkin_make
    source ~/catkin_ws/devel/setup.bash

3. VINS-Mask on Public datasets

Download EuRoC MAV Dataset. We only use one camera.

3.1 Run feature extraction service.

    # If you have an Anaconda environment
    conda activate YOUR_ENV
    # Run feature extraction service
    cd ~/catkin_ws/src/VINS-Mask/deep_feature
    python feat_service.py

3.2 Launch the vins_estimator, rviz and play the bag file respectively. Take MH_01 for example

    roslaunch vins_estimator euroc_no_extrinsic_param.launch
    roslaunch vins_estimator vins_rviz.launch
    rosbag play YOUR_PATH_TO_DATASET/MH_01_easy.bag 

3.3 More helpful tutorials can be found in VIMS-Mono.

4. Citation

@INPROCEEDINGS{vinsmask,
  author={Sun, Jiayuan and Song, Fangwei and Ji, Luping},
  booktitle={2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)}, 
  title={VINS-Mask: A ROI-mask Feature Tracker for Monocular Visual-inertial System}, 
  year={2022}
}

5. Acknowledgements

Thanks for VINS-Mono.