Uploaded August 2026 | Updated September 2026, 8 hours ago
This is the project video for Jonas Eggenberger's bachelor's thesis in the computer science department at the Bern University of Applied Sciences in Spring 2026.
3D Gaussian Splatting was introduced by Kerbl et al. as a method for real-time radiance field rendering and has since become an important alternative to implicit neural radiance field approaches. Instead of representing a scene through a neural network that must be evaluated during rendering, Gaussian Splatting uses an explicit set of anisotropic three-dimensional Gaussian primitives. Each primitive stores geometric and appearance-related parameters and can be projected directly into image space. This explicit representation is especially interesting for applications in computer graphics, computer vision, and virtual reality. A reconstructed scene can be rendered from novel viewpoints with high visual quality while keeping the rendering process comparatively direct and interpretable. This makes Gaussian Splatting suitable for interactive visualization and for systems where the relationship between scene parameters, projection, rasterization, and image formation should remain understandable.
This is the project video for Jonas Eggenberger's bachelor's thesis in the computer science department at the Bern University of Applied Sciences in Spring 2026.
3D Gaussian Splatting was introduced by Kerbl et al. as a method for real-time radiance field rendering and has since become an important alternative to implicit neural radiance field approaches. Instead of representing a scene through a neural network that must be evaluated during rendering, Gaussian Splatting uses an explicit set of anisotropic three-dimensional Gaussian primitives. Each primitive stores geometric and appearance-related parameters and can be projected directly into image space. This explicit representation is especially interesting for applications in computer graphics, computer vision, and virtual reality. A reconstructed scene can be rendered from novel viewpoints with high visual quality while keeping the rendering process comparatively direct and interpretable. This makes Gaussian Splatting suitable for interactive visualization and for systems where the relationship between scene parameters, projection, rasterization, and image formation should remain understandable.
