About The Project

Our VAMR VO pipeline project :)

Project report

Link to project report PDF

Getting Started

To get a local copy up and running follow these steps.

Prerequisites

1. Install conda
2. Clone this repo

Installation

1. Navigate to the project directory and run conda env create -f environment.yml
2. Run conda activate vamr_vo

Running

1. Navigate to the VAMR_VO directory
2. Choose the dataset in params/params.py. If not using the shared dataset, make sure you have the chosen dataset in your local_data folder.
3. Set params for dataset. In particular, set DISTRIBUTE_KEYPOINTS = True for Malaga, and DISTRIBUTE_KEYPOINTS = False for Kitti to speed it up.
4. Run python visual_odometry/main.py

Links

• Subject website
• Project FAQ website

Datasets

The shared data folder contains the small/easy dataset. The other larger datasets should be stored locally in a top level folder local_data/. This is in the .gitignore.

Development Guidelines

• Do work in your own branch, with the prefix of your github id followed by feature name. Ex: git checkout -b 'kappibw/project-skeleton'
• Annotate all np arrays with the expected shape/dimensions in inline comments.
• Add someone else as reviewer on a PR before merging to master.
• Document your work in the draft report as you go, use it as a scratchbook.