Vision-based odometry allows a vehicle to localize itself robustly by using only a stream of images captured by a camera attached to the vehicle. Idea behind visual odometry is to track distinctive features in consecutive images and use the displacement of these features to calculate the camera or robot's motion relative to its surroundings.
The dataset used is KITTI dataset. Dataset Link
cd <path to clone the repo>
git clone https://github.com/ayushabrol13/Visual-Odometry.git
cd Visual-Odometry
pip install virtualenv
virtualenv venv
source venv/bin/activate
pip install -r requirements.txt
python3 monoVisualOdometry.py
python3 featureTracking.py
python3 stereoVisualOdometry.py
To run the complete code, run through the cells of ipynb notebook CV_VO.ipynb
Devyani Gorkar B20ME027
Ayush Abrol B20AI052
Anjali Agarwal B20AI051
Aryan Tiwari B20AI056

