GaussianGPT is a research implementation for autoregressive generation of 3D scenes represented with Gaussian Splatting. It supports scene generation, completion, and outpainting through a locally run Python-based model intended for computer vision researchers and developers.
GaussianGPT sits in PulseGate's Data science & ML workbench category. It focuses on creating and editing 3D Gaussian Splatting scenes without manually constructing every scene element. GaussianGPT is an open-source project aimed at 3D computer vision researchers and developers. It ships for the command line, and it can be self-hosted.
Nicolas von Luetzow builds and maintains GaussianGPT. Key capabilities include 3D scene generation, scene completion, and scene outpainting.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match