DynamicPointCloudSubjectiveTest

This repository contains a dynamic point cloud (Volumetric video) subjective test environment.

Required Software

Unity 2019.3.19f1.

SteamVR for HTC Vive Pro.

Content Preparation

The environment currently supports PLY format with plain ASCII, binary little-endian, and binary big-endian, utilizing miniply A fast and easy-to-use library for parsing PLY files as the importer. For optimal performance, it is essential that all point cloud frames (.PLY files) are in binary format to ensure a playable framerate.

Setup the test

To get started, follow these steps:

1. Clone this repository to your computer.
2. Copy all point cloud sequences prepared in Content Preparation into the folder Assets\PointCloud with the following structure:

─── PointCloud
   └─── loot
   └─── redandblack
   └─── soldier
   └─── longdress
       ├─── representation1
       ├─── representation2
       ├─── representation3
       ├─── representation4
       └─── representation5
          ├─── longdress0000.ply
          ├─── longdress0001.ply
          ├─── longdress0002.ply
          ...
          └─── longdress0300.ply

3. Recheck Unity Project setting and build setting:
   • Single Pass Instanced rendering enabled.
   • All scences are included in the build setting.

Test reproduction

• To reconduct the Impacts of Temporal Quality Variation test, play the scene Assets\Scene\VersionSwitch\Training.unity and follow the instructions.
• To reconduct the Impact of Stalling test, play the scene Assets\Scene\Stall\longdressStall.unity and follow the instructions.

Citation

@inproceedings{nguyen2024modeling,
title = {Modeling User Quality of Experience in Adaptive Point Cloud Streaming},
author = {Nguyen, Duc V and Quang, Nguyen Long and Hien, Tran Thuy and Huyen, Nguyen Ngoc and Huong, Truong Thu and Nam, Pham Ngoc},
booktitle = {Proc. IEEE International Symposium on Multimedia},
pages = {1--6},
month = {Dec.},
year = {2024},
address = {Tokyo,Japan}
}