About

This is fork from VINS-Fusion-gpu for OpenCV 4 (and some fix).

VINS-Fusion-gpu is a version of VINS-Fusion with GPU acceleration.

Dependencies

• Ceres Solver
• ROS
• OpenCV
• OpenCV bridge

Ceres Solver

Installation.

Download latest stable release. Test it on ceres-solver-1.14 and ceres-solver-2.0.

Dependencies

sudo apt-get install -y libgoogle-glog-dev libatlas-base-dev libsuitesparse-dev

Build

mkdir ceres-build
cd ceres-build
cmake -D BUILD_TESTING=OFF \
      -D BUILD_EXAMPLES=OFF \
      ../ceres-solver-2.0.0
sudo make install

OpenCV bridge

This is require ROS and OpenCV bridge for OpenCV 4.

cd ~/catkin_ws/src
git clone https://github.com/ros-perception/vision_opencv

In CMakeLists.txt change python version (if necessary):

nano vision_opencv/cv_bridge/CMakeLists.txt

find_package(Boost REQUIRED python37) -> find_package(Boost REQUIRED python3)

In module.hpp add define NUMPY_IMPORT_ARRAY_RETVAL (if necessary):

nano vision_opencv/cv_bridge/src/module.hpp

#include <numpy/ndarrayobject.h>
#define NUMPY_IMPORT_ARRAY_RETVAL NULL

Build OpenCV bridge:

cd ~/catkin_ws
catkin_make

Build

Build is same as VINS-Fusion-gpu and VINS-Fusion.

Dependencies

sudo apt-get install ros-melodic-tf ros-melodic-image-transport

Build

Clone the repository and catkin_make:

cd ~/catkin_ws/src
git clone https://github.com/IOdissey/VINS-Fusion-GPU.git
cd ../
catkin_make
source ~/catkin_ws/devel/setup.bash

Usage

In the config file, there are two parameters for gpu acceleration.

GPU for GoodFeaturesToTrack (0 - off, 1 - on)

use_gpu: 1

GPU for OpticalFlowPyrLK (0 - off, 1 - on)

use_gpu_acc_flow: 1

If your GPU resources is limitted or you want to use GPU for other computaion. You can set

use_gpu: 1
use_gpu_acc_flow: 0

If your other application do not require much GPU resources (recommanded)

use_gpu: 1
use_gpu_acc_flow: 1

There are two parameters for config OpticalFlowPyrLK (pyramid level and window size):

lk_n: 3
lk_size: 21