We’ve been closely following the rapid progress in radiance fields, especially the evolution of Gaussian Splatting. This emerging visualization technique makes it possible to generate 3DGS (3D Gaussian Splatting) scenes and helps overcome many of the material property limitations found in traditional photogrammetry.
Since the technology is advancing quickly, we are taking a measured approach. For now, we’re experimenting with datasets originally captured for photogrammetry. It is encouraging to see that existing photogrammetry images can be reprocessed into formats like Gaussian Splats. This opens up exciting opportunities to reimagine our large archive of legacy photogrammetry data in contemporary formats such as Neural Radiance Fields (NeRFs) and Gaussian Splats.
The 3DGS model shown below was created using photogrammetry images of an insect specimen from the National Museum of Natural History.