/20-GROOT

Code for GROOT: A Real-Time Streaming System of High-Fidelity Volumetric Videos

Primary LanguageC


GROOT

Table of Contents
  1. About The Project
  2. Getting Started
  3. Contact
  4. Citation
  5. Acknowledgments

About The Project

Official source code for project "GROOT: A Real-time Streaming System of High-Fidelity Volumetric Videos" (MobiCom'20)

(back to top)

Built With

(back to top)

Getting Started

Currently only offline testing only. Online streaming code coming soon.

Dataset

Download dataset from 8i Voxelized Full Bodies

Server

Test encode/decode Groot on server

Prerequisites

Tested on Ubuntu 18.04

  • Install PCL
apt install -y libeigen3-dev libflann-dev libvtk6-qt-dev libpcap-dev  libboost-all-dev freeglut3-dev

wget https://github.com/PointCloudLibrary/pcl/archive/pcl-1.9.1.tar.gz
tar zvfx pcl-1.9.1.tar.gz

cd pcl-pcl-1.9.1
mkdir build && cd build
cmake ..
make -j2
make -j2 install
  • Install TurboJpeg
apt install libturbojpeg libturbojpeg0-dev
  • Install Zstandard
git clone https://github.com/facebook/zstd.git
cd zstd
make
make install

Test Encoding/Decoding

  • Encoding

    Make sure to check the scale of the data and set the smallest voxel size accordingly. For example, 8i dataset uses millimeter scale for the x, y, z coordinates (value range between 1~1000). The smallest voxel size should be set to 1. For the ease of rendering, I used the meter scale so I used a scaling factor of 0.001 and set the smallest voxel size as 0.001. The default value is set as scaling factor of 1 and smallest voxel size as 0.001. Change the main.cpp as needed.

# build
cd GrootServer/pdtree-encoding
mkdir build
cd build
cmake ..
make 

# refer to GrootServer/pdtree-encoding/run.sh for example
./server <dataset name: 8i/panoptic/custom> path/to/ply/files path/to/mortoncode path/to/output/folder 0

Mobile

iOS-based Groot decoder and renderer

(back to top)

Contact

Kyungjin Lee - @kyungjinleeee - kjlee818@gmail.com

Project Link: https://kyungjin-lee.github.io/groot/

(back to top)

Citation

If you find our work useful, please cite our paper below!

@inproceedings{10.1145/3372224.3419214,
author = {Lee, Kyungjin and Yi, Juheon and Lee, Youngki and Choi, Sunghyun and Kim, Young Min},
title = {GROOT: A Real-Time Streaming System of High-Fidelity Volumetric Videos},
year = {2020},
isbn = {9781450370851},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3372224.3419214},
doi = {10.1145/3372224.3419214},
booktitle = {Proceedings of the 26th Annual International Conference on Mobile Computing and Networking},
articleno = {57},
numpages = {14},
keywords = {point cloud, mobile augmented reality, virtual reality, volumetric video, video streaming},
location = {London, United Kingdom},
series = {MobiCom '20}
}

(back to top)

Acknowledgments

(back to top)