Yuzhou Ji, He Zhu, Junshu Tang, Wuyi Liu, Zhizhong Zhang...
TL;DR快速多语种查询 3D 场景的实时系统
Abstract
The semantically interactive radiance field has always been an appealing task
for its potential to facilitate user-friendly and automated real-world 3D scene
understanding applications. However, it is a challenging task to achieve high
quality, efficiency and zero-shot ability at the s
RT-GS2 是第一个能够实现实时泛化的 3D 高斯渲染中语义分割方法,通过提取独立于视角的 3D 高斯特征并采用新颖的视角相关 / 独立特征融合,能够在不同视角下提升语义一致性,超过同领域最先进方法的语义分割质量达到 8.01%的提高,并以每秒 27.03 帧的实时性能,相比现有方法提速达 901 倍,这一工作在引入光辐射场 3D 高斯表达领域的语义分割上代表了重大进展。