This repository is a MATLAB implementation as part of the Master Thesis Project Sparse Stereo Visual Odometry with Local Non-Linear Least-Squares Optimization for Navigation of Autonomous Vehicles.
The code has been set up to be used with the KITTI Vision Benchmark Suit.
The code has been written tested on MATLAB R2019a and depends on the following toolboxes:
- Computer Vision Toolbox
- Image Processing Toolbox
- Optimization Toolbox
Also, the code uses OpenCV 3.4.1 libraries which are implementen in MEX files.
The software was tested on a laptop Intel(R) Core(TM) i7-6500U CPU @ 2.50GHz 2.60 GHz and 12 GB RAM.
- Clone the repository using the following command:
git clone https://github.com/FabianAC07/Stereo-Visual-Odometry-with-KITTI-Vision-Benchmark
-
Install mexopencv
-
Import the dataset to the folder
dataset. You can request a download for KITTI "data_odometry_gray.zip" dataset here -
Change the corresponding paramters in the parameters
parameters.mfile according to your needs -
Run the script
main.m -
Depending on your choice in the parametes file, you might get a plot of the Visual Odometry estimation during sequence processing
-
Once the process is done, a new file
/resultswill be created. It will host all the outputs that you request in the parameters file, which can include:
- Video of the sequence
- Command Window Output
- Workspace Output for further post-processing
- Use the script plot_results.m to plot the results from Workspace Output file.
The following video is a demo of the plot output while processing...
Video: https://www.youtube.com/watch?v=NaQ8bB9eXo4
For details on the implemenation and use of this software please refer to the M.A.Sc. Thesis Sparse Stereo Visual Odometry with Local Non-Linear Least-Squares Optimization for Navigation of Autonomous Vehicles.
This software is under GNU General Public License v3.0 License.
If you use this software in an academic work, please cite:
E. F. Aguilar Calzadillas, "Sparse Stereo Visual Odometry with Local Non-Linear Least-Squares Optimization for Navigation of Autonomous Vehicles", M.A.Sc. Thesis, Depart. of Mech. and Aero. Eng., Carleton University, Ottawa ON, Canada, 2019.