ONL | Efficient Gaussian Splatting for Virtual Reality Applications
© Jeremias Michael Kilian und Jakob Béla Ruckel
Project information
submitted by
Adrian Kreskowski
Co-Authors
Jeremias Michael Kilian, Jakob Béla Ruckel
Mentors
Gareth Rendle, Adrian Kreskowski, Simon Schneegans, Prof. Dr. Bernd Fröhlich
Faculty:
Media
Degree programme:
Computer Science and Media (Bachelor of Science (B.Sc.)),
Computer Science and Media (english) (Master of Science (M.Sc.)),
Informatik (Bachelor of Science (B.Sc.))
Type of project presentation
Presentation
Semester
Wintersemester 2024/25
- Bauhausstraße 11
(N104 - Virtual Reality Labor)
Freitag, 17 - 22 Uhr & Samstag 12 - 15 Uhr
attractive to children
Project description
In the project Efficient Gaussian Splatting for Virtual Reality Applications, our project students focused on the analysis and hardware-dependent optimization of modern real-time rendering algorithms for the visualization of realistic models rendered using 3D Gaussian Splatting.
At summaery 2025, Jakob and Jeremias will present an optimized rendering plugin for Unity that allows even large Gaussian Splatting-based datasets, such as various rooms of the Haus am Horn, to be explored efficiently in terms of computation and with visually appealing results.
In addition to the practical demo, all project participants look forward to answering questions from interested visitors about rendering details, particularly regarding optimized visualization of Gaussian Splats on both desktop and mobile hardware!
Email: adrian.kreskowski[at]uni-weimar.de