princeton-vl/DPVO_Docker

Docker for Deep Patch Visual Odometry

This repository produces a docker container for our paper:

Deep Patch Visual Odometry (Source Code)
Zachary Teed, Lahav Lipson, Jia Deng

@article{teed2022deep,
  title={Deep Patch Visual Odometry},
  author={Teed, Zachary and Lipson, Lahav and Deng, Jia},
  journal={arXiv preprint arXiv:2208.04726},
  year={2022}
}

Install Dependencies

You can install Docker here

Also add the nvidia-docker repository

curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | \
  sudo tee /etc/apt/sources.list.d/nvidia-docker.list

Install the Nvidia container/docker toolkits

sudo apt-get update
sudo apt-get install -y nvidia-container-toolkit nvidia-docker2
sudo systemctl restart docker

Container Setup

Build the docker container

sudo ./build_container.sh <cuda-version>.1

Example: If nvidia-smi shows CUDA Version: 11.6 then run sudo ./build_container.sh 11.6.1

Start an interactive shell

sudo docker exec -it dpvo bash

Install DPVO and Pangolin

conda activate dpvo
./install_dpvo.sh

You should now be able to run DPVO

./download_models_and_data.sh
python demo.py --imagedir=movies/IMG_0494.MOV --calib=calib/iphone.txt --stride=5 --viz