About

Docker Container of torch-ngp to fix all dependency issues. With added Python script to go straight from video to NeRF with one command.

Container Setup and Use

First, Clone this repo

git clone https://github.com/mustafahamoody/torch_ngp_container

Change working directory to the docker folder in this repo:

cd torch_ngp_container/docker/

Build and run container (detached) with correct dependencies and mounted folder for data:

docker compose up -d

Enter running container:

docker-compose exec torch-ngp bash

Using env_creation.py -- Inside Container

To create an environment for the first time (From video or photos all the way to NeRF)

python env_create.py --run

• You will be prompted to name the environment you want to create, select the data type and provide the path to the data
• Videos or photos should be added to the parent folder, torch_ngp, (from your host machine) and when prompted, for the path type the file name (eg box.mp4)

To create an environment from a data folder (Photos already processed through COLMAP)

python env_create.py --train

To view NeRF environment in GUI

python env_create.py --view

To visualise the created environment using the GUI you must connect your host machine display to the docker container using X11 To Set this up run the following commands On your host machine:

** For Linux Machines: **

# 1. Install X11 Server
sudo apt-get install xorg openbox

# 2. Allow Docker to access your X server
xhost +local:docker

Citation

If you use this work, please include the following citations

Original NeRF authors:

@misc{mildenhall2020nerf,
    title={NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis},
    author={Ben Mildenhall and Pratul P. Srinivasan and Matthew Tancik and Jonathan T. Barron and Ravi Ramamoorthi and Ren Ng},
    year={2020},
    eprint={2003.08934},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Instant-NGP:

@article{mueller2022instant,
    title = {Instant Neural Graphics Primitives with a Multiresolution Hash Encoding},
    author = {Thomas M\"uller and Alex Evans and Christoph Schied and Alexander Keller},
    journal = {arXiv:2201.05989},
    year = {2022},
    month = jan
}

torch-ngp:

@misc{torch-ngp,
    Author = {Jiaxiang Tang},
    Year = {2022},
    Note = {https://github.com/ashawkey/torch-ngp},
    Title = {Torch-ngp: a PyTorch implementation of instant-ngp}
}

torch_ngp_container:

@misc{torch_ngp_container,
    Author = {Mustafa Hamoody},
    Year = {2025},
    Note = {https://github.com/mustafahamoody/torch_ngp_container},
    Title = {torch_ngp_container: Docker container of torch-ngp with easier NeRF creation}