GS Quality Assessment (GSQA)
收藏资源简介:
GS Quality Assessment (GSQA)数据集由密苏里大学堪萨斯城分校科学与工程学院和上海交通大学媒体创新合作中心共同创建,旨在评估3D高斯喷射(GS)的质量。该数据集包含120个样本,涵盖静态和动态GS内容。数据集的创建过程涉及使用GGSC方法对GS样本进行分割、图构建、信号残差矩阵生成、量化和编码。GSQA数据集的应用领域主要集中在3D场景重建和压缩技术中,旨在解决高斯喷射数据的传统压缩方法的研究空白问题。
The GS Quality Assessment (GSQA) dataset was jointly created by the School of Science and Engineering at the University of Missouri-Kansas City and the Media Innovation Center at Shanghai Jiao Tong University. It is designed to evaluate the quality of 3D Gaussian Spray (GS). The dataset contains 120 samples, encompassing both static and dynamic GS content. The creation process of the dataset involves segmenting GS samples using the GGSC method, constructing graphs, generating signal residual matrices, quantification, and encoding. The application of the GSQA dataset is primarily focused on 3D scene reconstruction and compression techniques, aiming to address the research gap in traditional compression methods for Gaussian Spray data.
GGSC数据集概述
数据集简介
- 数据集名称:A Benchmark for Gaussian Splatting Compression and Quality Assessment Study
- 研究领域:3D高斯泼溅压缩与质量评估
- 主要用途:提供高斯泼溅(Gaussian Splatting, GS)压缩与质量评估的基准测试
数据集内容
- 示例数据下载地址:https://drive.google.com/file/d/1QYOBdPcS0M2YXMbV4G_N7-_kKy9htpaw/view?usp=sharing
- GSQA数据集下载地址:
- OneDrive:https://1drv.ms/f/c/3dbe3858aa085846/EjN7TaKq5QBNhgE83zV5plIB5RcFIUmlV0VCvzsXpFp8pQ?e=5aGyg8
- 百度网盘:https://pan.baidu.com/s/14zJyYmbj8pMlsPqnBMBuzQ 密码:6njd
脚本运行结果
运行脚本后生成以下文件:
xxx_geo.bin:GS中心的比特率xxx_attr.bin:GS属性的比特率xxx_rec.ply:压缩后的重建GS样本xxx.json:压缩参数xxx_time.txt:压缩时间日志
引用信息
bibtex @inproceedings{yang2024GGSC, author = {Yang, Qi and Yang, Kaifa and Xing, Yuke and Xu, Yiling and Li, Zhu}, title = {A Benchmark for Gaussian Splatting Compression and Quality Assessment Study}, year = {2024}, isbn = {9798400712739}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3696409.3700172}, doi = {10.1145/3696409.3700172}, booktitle = {Proceedings of the 6th ACM International Conference on Multimedia in Asia}, articleno = {12}, numpages = {8}, keywords = {3D Gaussian Splatting, Compression, Quality Assessment}, series = {MMAsia 24} }
环境要求
- Python库:
- arithmetic_compressor
- scipy
- matplotlib

- 1A Benchmark for Gaussian Splatting Compression and Quality Assessment Study密苏里大学堪萨斯城分校科学与工程学院,上海交通大学媒体创新合作中心 · 2024年



