3D Gaussian Splatting is a new algorithm for synthesising photo-realistic 3D scenes from 2D images. This algorithm has been developed in 2023 by INRIA in Sophia Antipolis. In just a few months, it has become the reference in the field. This solution is broken down into 2 parts: a tool for training scenes with Pytorch based on AI algorithms and a 3D Gaussian Splatting scene viewer developed in C++/CUDA (named SIBR_gaussianViewer).
As part of the activities of the Orange research program on volumetric video, the team ported the SIBR_gaussianViewer viewer to OpenXR which is an open standard to support most virtual reality headsets on the market (Quest, HTC Vive…). This viewer was demonstrated at Open Tech Days 2023 and received strong interest from the public. Visitors could immerse themselves in sports scenes such as high jumping or breakdancing.
At the beginning of 2024, this development has been upstreamed to the INRIA SIBR_core official repository as an Orange opensource contribution. This contribution highlights our strong commitment on this subject and consolidates our close relationship with the INRIA’s team. The project can be found on INRIA’s GitLab.