blancocd/Geometrically_Accurate_FeedForward_Gaussian_Splatting_for_Unbounded_Scenes

Master's Thesis: Geometrically Accurate Feed‐Forward Gaussian Splatting for Unbounded Scenes

César Díaz Blanco

Supervisors: Zehao Yu · Haofei Xu

Examiners: Andreas Geiger · Gerard Pons-Moll

Thesis | Project Page | Code

This thesis enables feed-forward Gaussian splatting models to achieve geometrically accurate 3D representations of unbounded scenes. Our method performs on par with costly per-scene optimization and surpasses state-of-the-art depth estimation models.