arrow
返回

Optimized wavelet-based texture representation and streaming for GPU texture mapping

delete2017-02-15
delete3
PRE
AI
B
Bob Andries *
J
Jan Lemeire
A
Adrian Munteanu
DOI:10.1007/s11042-017-4433-3delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Because of the ever increasing resolution of consumer displays, high quality real-time 3D rendering applications require large amounts of 2D texture data, which in turn require large amounts of storage, memory and bandwidth. One of the major advantages of using compressed texture data is the significant decrease in bandwidth requirements, decreasing texture loading times. Additionally, instead of preloading all the potentially required information, texture data can be streamed progressively at run-time, which is a very common scenario in web-based applications, games and large virtual environments. This paper proposes a new texture streaming system, utilizing both pre-computed and run-time scene analysis, camera prediction and wavelet-based texture compression to provide maximal visual quality within bandwidth and real-time constraints. Texture decoding is done on the GPU, saving both on streaming bandwidth and GPU memory. The pre-computed scene analysis system can perform the run-time analysis at virtually no performance cost in multi-threaded systems. The proposed solution can easily be plugged into existing classical texture mapping solutions, as it features drop-in replacement shaders and re-uses existing render facilities. The wavelet-based streaming system results in PSNR improvements of up to 2 dB when compared to a DXT1-based streaming system.
Keyword:
Texture mapping
Texture streaming
Wavelet transform
Texture compression
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Multimedia Tools and Applications 封面图
Multimedia Tools and Applications
IF:
3
论文数:
2.0W
被引数:
3.2W

机构

V
Vrije Universiteit Brussel
学者数:
1.4W
论文数: 1.3W
被引数: 129
引用论文

引用论文

HTML5-based system for interoperable 3D digital home applications
err2013-06-05
err8
PREAI
errZorrilla, Mikel; Martin, Angel; Sanchez, Jairo R.; Tamayo, Inigo; Olaizola, Igor G.
err分享
err收藏
err分享
err收藏
学者 查看更多内容